From 9ee0eeba6ed2df09066b6b926867279e4db7fac5 Mon Sep 17 00:00:00 2001 From: michalovadek <46047818+michalovadek@users.noreply.github.com> Date: Mon, 15 Mar 2021 11:47:54 +0100 Subject: [PATCH] version 0.3.5 --- DESCRIPTION | 3 +- NEWS.md | 2 +- R/elx_make_query.R | 5 +- README.md | 17 +- doc/eurlexpkg.R | 28 +- doc/eurlexpkg.Rmd | 41 +- doc/eurlexpkg.html | 490 ++++++++++-------- docs/404.html | 2 +- docs/articles/eurlexpkg.html | 85 ++- .../figure-html/unnamed-chunk-9-1.png | Bin 9719 -> 56081 bytes .../figure-html/wordcloud-1.png | Bin 30282 -> 30617 bytes docs/articles/index.html | 2 +- docs/authors.html | 2 +- docs/index.html | 23 +- docs/news/index.html | 4 +- docs/pkgdown.yml | 2 +- docs/reference/elx_council_votes.html | 2 +- docs/reference/elx_curia_list.html | 14 +- docs/reference/elx_fetch_data.html | 2 +- docs/reference/elx_label_eurovoc.html | 2 +- docs/reference/elx_make_query.html | 34 +- docs/reference/elx_parse_xml.html | 2 +- docs/reference/elx_run_query.html | 6 +- docs/reference/index.html | 2 +- man/elx_make_query.Rd | 4 +- vignettes/eurlexpkg.Rmd | 32 +- 26 files changed, 492 insertions(+), 314 deletions(-) diff --git a/DESCRIPTION b/DESCRIPTION index 6cc2b3c..9e59232 100644 --- a/DESCRIPTION +++ b/DESCRIPTION @@ -32,7 +32,6 @@ Suggests: tidytext, wordcloud, purrr, - ggplot2, - glue + ggplot2 URL: https://michalovadek.github.io/eurlex/ VignetteBuilder: knitr diff --git a/NEWS.md b/NEWS.md index 2c107f5..99b3bda 100644 --- a/NEWS.md +++ b/NEWS.md @@ -4,7 +4,7 @@ - it is now possible to select all resource types available with `elx_make_query(resource_type = "any")`. Since there are nearly 1 million CELEX codes, use with discretion and expect long execution times - results can be restricted to a particular directory code with `elx_make_query(directory = "18")` (directory code "18" denotes Common Foreign and Security Policy) -- results can be restricted to a particular sector with `elx_make_query(sector = 2)` (sector code 3 denotes EU international agreements) +- results can be restricted to a particular sector with `elx_make_query(sector = 2)` (sector code 2 denotes EU international agreements) ## Minor changes diff --git a/R/elx_make_query.R b/R/elx_make_query.R index b72251b..756e983 100644 --- a/R/elx_make_query.R +++ b/R/elx_make_query.R @@ -1,8 +1,8 @@ -#' Create SPARQL quries +#' Create SPARQL queries #' #' Generates pre-defined or manual SPARQL queries to retrieve document ids from Cellar. #' List of available resource types: http://publications.europa.eu/resource/authority/resource-type . -#' Note that not all resource types are compatible with the pre-defined query. +#' Note that not all resource types are compatible with default parameter values. #' #' @importFrom magrittr %>% #' @@ -46,6 +46,7 @@ elx_make_query <- function(resource_type = c("directive","regulation","decision" include_directory = FALSE, include_sector = FALSE, order = FALSE, limit = NULL){ + if (missing(resource_type)) stop("'resource_type' must be defined") if (!resource_type %in% c("any","directive","regulation","decision","recommendation","intagr","caselaw","manual","proposal","national_impl")) stop("'resource_type' must be defined") if (resource_type == "manual" & nchar(manual_type) < 2){ diff --git a/README.md b/README.md index d384a33..51d079c 100644 --- a/README.md +++ b/README.md @@ -26,11 +26,11 @@ For the moment, it is recommended to retrieve metadata one variable at a time. F 2. `dates <- elx_make_query("directive", include_date_transpos = TRUE) %>% elx_run_query()` 3. `ids %>% dplyr::left_join(lbs) %>% dplyr::left_join(dates)` -rather than `elx_make_query("directive", include_lbs = TRUE, include_date_transpos = TRUE)`. This approach should make it easier to understand the returned data frame(s), especially when some variables contain missing or duplicated data. +rather than `elx_make_query("directive", include_lbs = TRUE, include_date_transpos = TRUE)`. This approach should make it easier to understand the returned data frame(s), especially when some variables contain missing or duplicated data. Always keep an eye on whether the `work` and `celex` columns identify rows uniquely or not. One of the main contributions of the SPARQL requests is that we obtain a comprehensive list of identifiers that we can subsequently use to obtain more data relating to the document in question. While the results of the SPARQL queries are useful also for webscraping (with the `rvest` package), the function `elx_fetch_data()` enables us to fire GET requests to retrieve data on documents with known identifiers (including Cellar URI). The function currently enables downloading the title and the full text of a document in all available languages. -See the [vignette](https://michalovadek.github.io/eurlex/articles/eurlexpkg.html) for a walkthrough on how to use the package. Check function documentation for most up-to-date overview of features. +See the [vignette](https://michalovadek.github.io/eurlex/articles/eurlexpkg.html) for a walkthrough on how to use the package. Check function documentation for most up-to-date overview of features. Example use cases are shown in this [paper](https://www.tandfonline.com/doi/full/10.1080/2474736X.2020.1870150). ## Cite Michal Ovádek (2021) Facilitating access to data on European Union laws, Political Research Exchange, 3:1, DOI: [10.1080/2474736X.2020.1870150](https://www.tandfonline.com/doi/full/10.1080/2474736X.2020.1870150) @@ -40,6 +40,19 @@ This package nor its author are in any way affiliated with the EU Publications O Please consider contributing to the maintanance and development of the package by reporting bugs or suggesting new features. +## Latest changes + +### eurlex 0.3.5 + +- it is now possible to select all resource types available with `elx_make_query(resource_type = "any")`. Since there are nearly 1 million CELEX codes, use with discretion and expect long execution times +- results can be restricted to a particular directory code with `elx_make_query(directory = "18")` (directory code "18" denotes Common Foreign and Security Policy) +- results can be restricted to a particular sector with `elx_make_query(sector = 2)` (sector code 2 denotes EU international agreements) + +- new feature: request date of court case submission `elx_make_query(include_date_lodged = TRUE)` +- new feature: request type of court procedure and outcome `elx_make_query(include_court_procedure = TRUE)` +- new feature: request directory code of legal act `elx_make_query(include_directory = TRUE)` +- `elx_curia_list()` has a new default parameter `parse = TRUE` which creates separate columns for `ecli`, `see_case`, `appeal` applying regular expressions on `case_info` + ## Useful resources Guide to CELEX numbers: https://eur-lex.europa.eu/content/tools/TableOfSectors/types_of_documents_in_eurlex.html diff --git a/doc/eurlexpkg.R b/doc/eurlexpkg.R index 488fba1..fb4df0b 100644 --- a/doc/eurlexpkg.R +++ b/doc/eurlexpkg.R @@ -17,27 +17,36 @@ results <- dirs %>% select(-force,-date) ## ----------------------------------------------------------------------------- query_dir %>% - glue::as_glue() # for nicer printing + cat() # for nicer printing elx_make_query(resource_type = "caselaw") %>% - glue::as_glue() + cat() elx_make_query(resource_type = "manual", manual_type = "SWD") %>% - glue::as_glue() + cat() ## ----------------------------------------------------------------------------- elx_make_query(resource_type = "directive", include_date = TRUE, include_force = TRUE) %>% - glue::as_glue() + cat() # minimal query: elx_make_query(resource_type = "directive") elx_make_query(resource_type = "recommendation", include_date = TRUE, include_lbs = TRUE) %>% - glue::as_glue() + cat() # minimal query: elx_make_query(resource_type = "recommendation") +## ----------------------------------------------------------------------------- +# request documents from directory 18 ("Common Foreign and Security Policy") +# and sector 3 ("Legal acts") + +elx_make_query(resource_type = "any", + directory = "18", + sector = 3) %>% + cat() + ## ----runquery, eval=FALSE----------------------------------------------------- # results <- elx_run_query(query = query_dir) # @@ -65,18 +74,14 @@ rec_eurovoc %>% ## ----eurovoctable------------------------------------------------------------- - eurovoc_lookup <- elx_label_eurovoc(uri_eurovoc = rec_eurovoc$eurovoc) print(eurovoc_lookup) - ## ----appendlabs--------------------------------------------------------------- - rec_eurovoc %>% left_join(eurovoc_lookup) - ## ----------------------------------------------------------------------------- eurovoc_lookup <- elx_label_eurovoc(uri_eurovoc = rec_eurovoc$eurovoc, alt_labels = TRUE, @@ -86,7 +91,6 @@ rec_eurovoc %>% left_join(eurovoc_lookup) %>% select(celex, eurovoc, labels) - ## ----getdatapur, message = FALSE, warning=FALSE, error=FALSE------------------ # the function is not vectorized by default elx_fetch_data(results$work[1],"title") @@ -117,7 +121,9 @@ dirs %>% ## ----------------------------------------------------------------------------- dirs %>% - ggplot(aes(x = as.Date(date), y = celex)) + + filter(!is.na(force)) %>% + mutate(date = as.Date(date)) %>% + ggplot(aes(x = date, y = celex)) + geom_point(aes(color = force), alpha = 0.1) + theme(axis.text.y = element_blank(), axis.line.y = element_blank(), diff --git a/doc/eurlexpkg.Rmd b/doc/eurlexpkg.Rmd index 7258dbd..05b1e3d 100644 --- a/doc/eurlexpkg.Rmd +++ b/doc/eurlexpkg.Rmd @@ -2,7 +2,7 @@ title: "eurlex: Retrieve data on European Union law in R" output: rmarkdown::html_vignette description: > - Retrieve efficiently tidy data on European Union law in R with + Retrieve data on European Union law in R with pre-defined SPARQL and REST queries. vignette: > %\VignetteIndexEntry{eurlex: Retrieve data on European Union law in R} @@ -29,6 +29,8 @@ The `eurlex` R package attempts to significantly reduce the overhead associated The `eurlex` package currently envisions the typical use-case to consist of getting bulk information about EU legislation into R as fast as possible. The package contains three core functions to achieve that objective: `elx_make_query()` to create pre-defined or customized SPARQL queries; `elx_run_query()` to execute the pre-made or any other manually input query; and `elx_fetch_data()` to fire GET requests for certain metadata to the REST API. +The package also contains largely self-explanatory functions for retrieving data on EU court cases (`elx_curia_list()`) and Council votes (`elx_council_votes()`) from outside Eur-Lex. + ## `elx_make_query()`: Generate SPARQL queries The function `elx_make_query` takes as its first argument the type of resource to be retrieved from the semantic database that powers Eur-Lex (and other publications) called Cellar. @@ -55,13 +57,13 @@ The choice of resource type is then reflected in the SPARQL query generated by t ```{r} query_dir %>% - glue::as_glue() # for nicer printing + cat() # for nicer printing elx_make_query(resource_type = "caselaw") %>% - glue::as_glue() + cat() elx_make_query(resource_type = "manual", manual_type = "SWD") %>% - glue::as_glue() + cat() ``` @@ -69,21 +71,33 @@ There are various ways of querying the same information in the Cellar database d The other arguments in `elx_make_query()` relate to additional metadata to be returned. The results include by default the [CELEX number](https://eur-lex.europa.eu/content/tools/TableOfSectors/types_of_documents_in_eurlex.html) and exclude corrigenda (corrections of errors in legislation). Other data needs to be opted into. Make sure to select ones that are logically compatible (e.g. case law does not have a legal basis). More options should be added in the future. -Note that availability of data for each variable has an impact on the results. The data frame returned by the query will be shrunken to the size of the variable with most missing data. It is recommended to always compare results from a desired query to a minimal query requesting only celex ids. +Note that availability of data for each variable might have an impact on the results. The data frame returned by the query might be shrunken to the size of the variable with most missing data. It is recommended to always compare results from a desired query to a minimal query requesting only celex ids. ```{r} elx_make_query(resource_type = "directive", include_date = TRUE, include_force = TRUE) %>% - glue::as_glue() + cat() # minimal query: elx_make_query(resource_type = "directive") elx_make_query(resource_type = "recommendation", include_date = TRUE, include_lbs = TRUE) %>% - glue::as_glue() + cat() # minimal query: elx_make_query(resource_type = "recommendation") ``` +You can also decide to not specify any resource types, in which case all types of documents will be returned. As there are over a million documents with a CELEX identifier, this is likely not efficient for a majority of users. But since version 0.3.5 it is possible to request documents belonging to a particular ["sector"](https://eur-lex.europa.eu/content/tools/TableOfSectors/types_of_documents_in_eurlex.html) or [directory code](https://eur-lex.europa.eu/browse/directories/legislation.html). + +```{r} +# request documents from directory 18 ("Common Foreign and Security Policy") +# and sector 3 ("Legal acts") + +elx_make_query(resource_type = "any", + directory = "18", + sector = 3) %>% + cat() +``` + Now that we have a query, we are ready to run it. ## `elx_run_query()`: Execute SPARQL queries @@ -135,20 +149,16 @@ rec_eurovoc %>% By default, the endpoint returns the EuroVoc concept codes rather than the labels (keywords). The function `elx_label_eurovoc()` needs to be called to obtain a look-up table with the labels. ```{r eurovoctable} - eurovoc_lookup <- elx_label_eurovoc(uri_eurovoc = rec_eurovoc$eurovoc) print(eurovoc_lookup) - ``` The results include labels only for unique identifiers, but with `dplyr::left_join()` it is straightforward to append the labels to the entire dataset. ```{r appendlabs} - rec_eurovoc %>% left_join(eurovoc_lookup) - ``` As elsewhere in the API, we can tap into the multilingual nature of EU documents also when it comes to the EuroVoc keywords. Moreover, most concepts in the thesaurus are associated with alternative labels; these can be returned as well (separated by a comma). @@ -161,7 +171,6 @@ eurovoc_lookup <- elx_label_eurovoc(uri_eurovoc = rec_eurovoc$eurovoc, rec_eurovoc %>% left_join(eurovoc_lookup) %>% select(celex, eurovoc, labels) - ``` ## `elx_fetch_data()`: Fire GET requests @@ -186,7 +195,7 @@ print(dir_titles) ``` -Note that text requests are by far the most time-intensive; requesting the full text for thousands of documents is liable to extend the run-time into hours. Currently, no method for downloading text in non-html/plain formats is implemented, which means pdf-only texts will be missing from the results.^[It is worth pointing out that the html and pdf contents of older case law differs. Whereas typically the html file is only going to contain a summary and grounds of a judgment, the pdf should also contain background to the dispute.] +Note that text requests are by far the most time-intensive; requesting the full text for thousands of documents is liable to extend the run-time into hours. Texts are retrieved from html by priority, but methods for pdfs and .docs are also implemented.^[It is worth pointing out that the html and pdf contents of older case law differs. Whereas typically the html file is only going to contain a summary and grounds of a judgment, the pdf should also contain background to the dispute.] The function even handles multi-document resources (by pasting them together). # Application @@ -213,7 +222,9 @@ Directives become naturally outdated with time. It might be all the more interes ```{r} dirs %>% - ggplot(aes(x = as.Date(date), y = celex)) + + filter(!is.na(force)) %>% + mutate(date = as.Date(date)) %>% + ggplot(aes(x = date, y = celex)) + geom_point(aes(color = force), alpha = 0.1) + theme(axis.text.y = element_blank(), axis.line.y = element_blank(), @@ -251,6 +262,6 @@ dirs_1970_title %>% I use term-frequency inverse-document frequency (tf-idf) to weight the importance of the words in the wordcloud. If we used pure frequencies, the wordcloud would largely consist of words conveying little meaning ("the", "and", ...). -This is an extremely basic application of the `eurlex` package. Much more sophisticated methods can be used to analyse both the content and metadata of European Union legislation. If the package is useful for your research, please consider citing it. +This is an extremely basic application of the `eurlex` package. Much more sophisticated methods can be used to analyse both the content and metadata of European Union legislation. If the package is useful for your research, please consider citing the [accompanying paper](https://www.tandfonline.com/doi/full/10.1080/2474736X.2020.1870150).^[Michal Ovádek (2021) Facilitating access to data on European Union laws, Political Research Exchange, 3:1, DOI: [10.1080/2474736X.2020.1870150](https://www.tandfonline.com/doi/full/10.1080/2474736X.2020.1870150)] diff --git a/doc/eurlexpkg.html b/doc/eurlexpkg.html index b4098ba..495b333 100644 --- a/doc/eurlexpkg.html +++ b/doc/eurlexpkg.html @@ -311,6 +311,7 @@

Introduction

The eurlex package

The eurlex package currently envisions the typical use-case to consist of getting bulk information about EU legislation into R as fast as possible. The package contains three core functions to achieve that objective: elx_make_query() to create pre-defined or customized SPARQL queries; elx_run_query() to execute the pre-made or any other manually input query; and elx_fetch_data() to fire GET requests for certain metadata to the REST API.

+

The package also contains largely self-explanatory functions for retrieving data on EU court cases (elx_curia_list()) and Council votes (elx_council_votes()) from outside Eur-Lex.

elx_make_query(): Generate SPARQL queries

The function elx_make_query takes as its first argument the type of resource to be retrieved from the semantic database that powers Eur-Lex (and other publications) called Cellar.

@@ -321,266 +322,306 @@

elx_make_query(): Generate SPARQL queries

Currently, it is possible to choose from among a host of resource types, including directives, regulations and even case law (see function description for the full list). It is also possible to manually specify a resource type from the eligible list.1

The choice of resource type is then reflected in the SPARQL query generated by the function:

query_dir %>% 
-  glue::as_glue() # for nicer printing
+  cat() # for nicer printing
 #> PREFIX cdm: <http://publications.europa.eu/ontology/cdm#>
-#>   PREFIX skos:<http://www.w3.org/2004/02/skos/core#>
-#>   PREFIX dc:<http://purl.org/dc/elements/1.1/>
-#>   PREFIX xsd:<http://www.w3.org/2001/XMLSchema#>
-#>   PREFIX rdf:<http://www.w3.org/1999/02/22-rdf-syntax-ns#>
-#>   PREFIX owl:<http://www.w3.org/2002/07/owl#>
-#>   select distinct ?work ?type ?celex where{ ?work cdm:work_has_resource-type ?type. FILTER(?type=<http://publications.europa.eu/resource/authority/resource-type/DIR>||
-#>   ?type=<http://publications.europa.eu/resource/authority/resource-type/DIR_IMPL>||
-#>   ?type=<http://publications.europa.eu/resource/authority/resource-type/DIR_DEL>) 
-#>  FILTER not exists{?work cdm:work_has_resource-type <http://publications.europa.eu/resource/authority/resource-type/CORRIGENDUM>} ?work cdm:resource_legal_id_celex ?celex. }
-
-elx_make_query(resource_type = "caselaw") %>% 
-  glue::as_glue()
-#> PREFIX cdm: <http://publications.europa.eu/ontology/cdm#>
-#>   PREFIX skos:<http://www.w3.org/2004/02/skos/core#>
-#>   PREFIX dc:<http://purl.org/dc/elements/1.1/>
-#>   PREFIX xsd:<http://www.w3.org/2001/XMLSchema#>
-#>   PREFIX rdf:<http://www.w3.org/1999/02/22-rdf-syntax-ns#>
-#>   PREFIX owl:<http://www.w3.org/2002/07/owl#>
-#>   select distinct ?work ?type ?celex where{ ?work cdm:work_has_resource-type ?type. FILTER(?type=<http://publications.europa.eu/resource/authority/resource-type/JUDG>||
-#>   ?type=<http://publications.europa.eu/resource/authority/resource-type/ORDER>||
-#>   ?type=<http://publications.europa.eu/resource/authority/resource-type/OPIN_JUR>||
-#>   ?type=<http://publications.europa.eu/resource/authority/resource-type/THIRDPARTY_PROCEED>||
-#>   ?type=<http://publications.europa.eu/resource/authority/resource-type/GARNISHEE_ORDER>||
-#>   ?type=<http://publications.europa.eu/resource/authority/resource-type/RULING>||
-#>   ?type=<http://publications.europa.eu/resource/authority/resource-type/JUDG_EXTRACT>||
-#>   ?type=<http://publications.europa.eu/resource/authority/resource-type/INFO_JUDICIAL>||
-#>   ?type=<http://publications.europa.eu/resource/authority/resource-type/VIEW_AG>||
-#>   ?type=<http://publications.europa.eu/resource/authority/resource-type/OPIN_AG>) ?work cdm:resource_legal_id_celex ?celex. }
-
-elx_make_query(resource_type = "manual", manual_type = "SWD") %>% 
-  glue::as_glue()
-#> PREFIX cdm: <http://publications.europa.eu/ontology/cdm#>
-#>   PREFIX skos:<http://www.w3.org/2004/02/skos/core#>
-#>   PREFIX dc:<http://purl.org/dc/elements/1.1/>
-#>   PREFIX xsd:<http://www.w3.org/2001/XMLSchema#>
-#>   PREFIX rdf:<http://www.w3.org/1999/02/22-rdf-syntax-ns#>
-#>   PREFIX owl:<http://www.w3.org/2002/07/owl#>
-#>   select distinct ?work ?type ?celex where{ ?work cdm:work_has_resource-type ?type.FILTER(?type=<http://publications.europa.eu/resource/authority/resource-type/SWD>) 
-#>  FILTER not exists{?work cdm:work_has_resource-type <http://publications.europa.eu/resource/authority/resource-type/CORRIGENDUM>} ?work cdm:resource_legal_id_celex ?celex. }
+#> PREFIX annot: <http://publications.europa.eu/ontology/annotation#> +#> PREFIX skos:<http://www.w3.org/2004/02/skos/core#> +#> PREFIX dc:<http://purl.org/dc/elements/1.1/> +#> PREFIX xsd:<http://www.w3.org/2001/XMLSchema#> +#> PREFIX rdf:<http://www.w3.org/1999/02/22-rdf-syntax-ns#> +#> PREFIX owl:<http://www.w3.org/2002/07/owl#> +#> select distinct ?work ?type ?celex where{ ?work cdm:work_has_resource-type ?type. FILTER(?type=<http://publications.europa.eu/resource/authority/resource-type/DIR>|| +#> ?type=<http://publications.europa.eu/resource/authority/resource-type/DIR_IMPL>|| +#> ?type=<http://publications.europa.eu/resource/authority/resource-type/DIR_DEL>) +#> FILTER not exists{?work cdm:work_has_resource-type <http://publications.europa.eu/resource/authority/resource-type/CORRIGENDUM>} OPTIONAL{?work cdm:resource_legal_id_celex ?celex.} } + +elx_make_query(resource_type = "caselaw") %>% + cat() +#> PREFIX cdm: <http://publications.europa.eu/ontology/cdm#> +#> PREFIX annot: <http://publications.europa.eu/ontology/annotation#> +#> PREFIX skos:<http://www.w3.org/2004/02/skos/core#> +#> PREFIX dc:<http://purl.org/dc/elements/1.1/> +#> PREFIX xsd:<http://www.w3.org/2001/XMLSchema#> +#> PREFIX rdf:<http://www.w3.org/1999/02/22-rdf-syntax-ns#> +#> PREFIX owl:<http://www.w3.org/2002/07/owl#> +#> select distinct ?work ?type ?celex where{ ?work cdm:work_has_resource-type ?type. FILTER(?type=<http://publications.europa.eu/resource/authority/resource-type/JUDG>|| +#> ?type=<http://publications.europa.eu/resource/authority/resource-type/ORDER>|| +#> ?type=<http://publications.europa.eu/resource/authority/resource-type/OPIN_JUR>|| +#> ?type=<http://publications.europa.eu/resource/authority/resource-type/THIRDPARTY_PROCEED>|| +#> ?type=<http://publications.europa.eu/resource/authority/resource-type/GARNISHEE_ORDER>|| +#> ?type=<http://publications.europa.eu/resource/authority/resource-type/RULING>|| +#> ?type=<http://publications.europa.eu/resource/authority/resource-type/JUDG_EXTRACT>|| +#> ?type=<http://publications.europa.eu/resource/authority/resource-type/INFO_JUDICIAL>|| +#> ?type=<http://publications.europa.eu/resource/authority/resource-type/VIEW_AG>|| +#> ?type=<http://publications.europa.eu/resource/authority/resource-type/OPIN_AG>) OPTIONAL{?work cdm:resource_legal_id_celex ?celex.} } + +elx_make_query(resource_type = "manual", manual_type = "SWD") %>% + cat() +#> PREFIX cdm: <http://publications.europa.eu/ontology/cdm#> +#> PREFIX annot: <http://publications.europa.eu/ontology/annotation#> +#> PREFIX skos:<http://www.w3.org/2004/02/skos/core#> +#> PREFIX dc:<http://purl.org/dc/elements/1.1/> +#> PREFIX xsd:<http://www.w3.org/2001/XMLSchema#> +#> PREFIX rdf:<http://www.w3.org/1999/02/22-rdf-syntax-ns#> +#> PREFIX owl:<http://www.w3.org/2002/07/owl#> +#> select distinct ?work ?type ?celex where{ ?work cdm:work_has_resource-type ?type.FILTER(?type=<http://publications.europa.eu/resource/authority/resource-type/SWD>) +#> FILTER not exists{?work cdm:work_has_resource-type <http://publications.europa.eu/resource/authority/resource-type/CORRIGENDUM>} OPTIONAL{?work cdm:resource_legal_id_celex ?celex.} }

There are various ways of querying the same information in the Cellar database due to the existence of several overlapping classes and identifiers describing the same resources. The queries generated by the function should offer a reliable way of obtaining exhaustive results, as they have been validated by the helpdesk of the Publication Office. At the same time, it is always possible there will be issues either on the query or the database side; please report any you encounter through Github.

The other arguments in elx_make_query() relate to additional metadata to be returned. The results include by default the CELEX number and exclude corrigenda (corrections of errors in legislation). Other data needs to be opted into. Make sure to select ones that are logically compatible (e.g. case law does not have a legal basis). More options should be added in the future.

-

Note that availability of data for each variable has an impact on the results. The data frame returned by the query will be shrunken to the size of the variable with most missing data. It is recommended to always compare results from a desired query to a minimal query requesting only celex ids.

+

Note that availability of data for each variable might have an impact on the results. The data frame returned by the query might be shrunken to the size of the variable with most missing data. It is recommended to always compare results from a desired query to a minimal query requesting only celex ids.

elx_make_query(resource_type = "directive", include_date = TRUE, include_force = TRUE) %>% 
-  glue::as_glue()
+  cat()
 #> PREFIX cdm: <http://publications.europa.eu/ontology/cdm#>
-#>   PREFIX skos:<http://www.w3.org/2004/02/skos/core#>
-#>   PREFIX dc:<http://purl.org/dc/elements/1.1/>
-#>   PREFIX xsd:<http://www.w3.org/2001/XMLSchema#>
-#>   PREFIX rdf:<http://www.w3.org/1999/02/22-rdf-syntax-ns#>
-#>   PREFIX owl:<http://www.w3.org/2002/07/owl#>
-#>   select distinct ?work ?type ?celex str(?date) ?force where{ ?work cdm:work_has_resource-type ?type. FILTER(?type=<http://publications.europa.eu/resource/authority/resource-type/DIR>||
-#>   ?type=<http://publications.europa.eu/resource/authority/resource-type/DIR_IMPL>||
-#>   ?type=<http://publications.europa.eu/resource/authority/resource-type/DIR_DEL>) 
-#>  FILTER not exists{?work cdm:work_has_resource-type <http://publications.europa.eu/resource/authority/resource-type/CORRIGENDUM>} ?work cdm:resource_legal_id_celex ?celex. ?work cdm:work_date_document ?date. ?work cdm:resource_legal_in-force ?force. }
-
-# minimal query: elx_make_query(resource_type = "directive")
-
-elx_make_query(resource_type = "recommendation", include_date = TRUE, include_lbs = TRUE) %>% 
-  glue::as_glue()
-#> PREFIX cdm: <http://publications.europa.eu/ontology/cdm#>
-#>   PREFIX skos:<http://www.w3.org/2004/02/skos/core#>
-#>   PREFIX dc:<http://purl.org/dc/elements/1.1/>
-#>   PREFIX xsd:<http://www.w3.org/2001/XMLSchema#>
-#>   PREFIX rdf:<http://www.w3.org/1999/02/22-rdf-syntax-ns#>
-#>   PREFIX owl:<http://www.w3.org/2002/07/owl#>
-#>   select distinct ?work ?type ?celex str(?date) ?lbs ?lbcelex where{ ?work cdm:work_has_resource-type ?type. FILTER(?type=<http://publications.europa.eu/resource/authority/resource-type/RECO>||
-#>                    ?type=<http://publications.europa.eu/resource/authority/resource-type/RECO_DEC>||
-#>                    ?type=<http://publications.europa.eu/resource/authority/resource-type/RECO_DIR>||
-#>                    ?type=<http://publications.europa.eu/resource/authority/resource-type/RECO_OPIN>||
-#>                    ?type=<http://publications.europa.eu/resource/authority/resource-type/RECO_RES>||
-#>                    ?type=<http://publications.europa.eu/resource/authority/resource-type/RECO_REG>||
-#>                    ?type=<http://publications.europa.eu/resource/authority/resource-type/RECO_RECO>||
-#>                    ?type=<http://publications.europa.eu/resource/authority/resource-type/RECO_DRAFT>) 
-#>  FILTER not exists{?work cdm:work_has_resource-type <http://publications.europa.eu/resource/authority/resource-type/CORRIGENDUM>} ?work cdm:resource_legal_id_celex ?celex. ?work cdm:work_date_document ?date. ?work cdm:resource_legal_based_on_resource_legal ?lbs.
-#>                    ?lbs cdm:resource_legal_id_celex ?lbcelex. }
-
-# minimal query: elx_make_query(resource_type = "recommendation")
+#> PREFIX annot: <http://publications.europa.eu/ontology/annotation#> +#> PREFIX skos:<http://www.w3.org/2004/02/skos/core#> +#> PREFIX dc:<http://purl.org/dc/elements/1.1/> +#> PREFIX xsd:<http://www.w3.org/2001/XMLSchema#> +#> PREFIX rdf:<http://www.w3.org/1999/02/22-rdf-syntax-ns#> +#> PREFIX owl:<http://www.w3.org/2002/07/owl#> +#> select distinct ?work ?type ?celex str(?date) ?force where{ ?work cdm:work_has_resource-type ?type. FILTER(?type=<http://publications.europa.eu/resource/authority/resource-type/DIR>|| +#> ?type=<http://publications.europa.eu/resource/authority/resource-type/DIR_IMPL>|| +#> ?type=<http://publications.europa.eu/resource/authority/resource-type/DIR_DEL>) +#> FILTER not exists{?work cdm:work_has_resource-type <http://publications.europa.eu/resource/authority/resource-type/CORRIGENDUM>} OPTIONAL{?work cdm:resource_legal_id_celex ?celex.} OPTIONAL{?work cdm:work_date_document ?date.} OPTIONAL{?work cdm:resource_legal_in-force ?force.} } + +# minimal query: elx_make_query(resource_type = "directive") + +elx_make_query(resource_type = "recommendation", include_date = TRUE, include_lbs = TRUE) %>% + cat() +#> PREFIX cdm: <http://publications.europa.eu/ontology/cdm#> +#> PREFIX annot: <http://publications.europa.eu/ontology/annotation#> +#> PREFIX skos:<http://www.w3.org/2004/02/skos/core#> +#> PREFIX dc:<http://purl.org/dc/elements/1.1/> +#> PREFIX xsd:<http://www.w3.org/2001/XMLSchema#> +#> PREFIX rdf:<http://www.w3.org/1999/02/22-rdf-syntax-ns#> +#> PREFIX owl:<http://www.w3.org/2002/07/owl#> +#> select distinct ?work ?type ?celex str(?date) ?lbs ?lbcelex ?lbsuffix where{ ?work cdm:work_has_resource-type ?type. FILTER(?type=<http://publications.europa.eu/resource/authority/resource-type/RECO>|| +#> ?type=<http://publications.europa.eu/resource/authority/resource-type/RECO_DEC>|| +#> ?type=<http://publications.europa.eu/resource/authority/resource-type/RECO_DIR>|| +#> ?type=<http://publications.europa.eu/resource/authority/resource-type/RECO_OPIN>|| +#> ?type=<http://publications.europa.eu/resource/authority/resource-type/RECO_RES>|| +#> ?type=<http://publications.europa.eu/resource/authority/resource-type/RECO_REG>|| +#> ?type=<http://publications.europa.eu/resource/authority/resource-type/RECO_RECO>|| +#> ?type=<http://publications.europa.eu/resource/authority/resource-type/RECO_DRAFT>) +#> FILTER not exists{?work cdm:work_has_resource-type <http://publications.europa.eu/resource/authority/resource-type/CORRIGENDUM>} OPTIONAL{?work cdm:resource_legal_id_celex ?celex.} OPTIONAL{?work cdm:work_date_document ?date.} OPTIONAL{?work cdm:resource_legal_based_on_resource_legal ?lbs. +#> ?lbs cdm:resource_legal_id_celex ?lbcelex. +#> OPTIONAL{?bn owl:annotatedSource ?work. +#> ?bn owl:annotatedProperty <http://publications.europa.eu/ontology/cdm#resource_legal_based_on_resource_legal>. +#> ?bn owl:annotatedTarget ?lbs. +#> ?bn annot:comment_on_legal_basis ?lbsuffix}} } + +# minimal query: elx_make_query(resource_type = "recommendation")
+

You can also decide to not specify any resource types, in which case all types of documents will be returned. As there are over a million documents with a CELEX identifier, this is likely not efficient for a majority of users. But since version 0.3.5 it is possible to request documents belonging to a particular “sector” or directory code.

+
# request documents from directory 18 ("Common Foreign and Security Policy")
+# and sector 3 ("Legal acts")
+
+elx_make_query(resource_type = "any",
+               directory = "18",
+               sector = 3) %>% 
+  cat()
+#> PREFIX cdm: <http://publications.europa.eu/ontology/cdm#>
+#>   PREFIX annot: <http://publications.europa.eu/ontology/annotation#>
+#>   PREFIX skos:<http://www.w3.org/2004/02/skos/core#>
+#>   PREFIX dc:<http://purl.org/dc/elements/1.1/>
+#>   PREFIX xsd:<http://www.w3.org/2001/XMLSchema#>
+#>   PREFIX rdf:<http://www.w3.org/1999/02/22-rdf-syntax-ns#>
+#>   PREFIX owl:<http://www.w3.org/2002/07/owl#>
+#>   select distinct ?work ?type ?celex where{
+#>     VALUES (?value)
+#>     { (<http://publications.europa.eu/resource/authority/fd_555/18>)
+#>       (<http://publications.europa.eu/resource/authority/dir-eu-legal-act/18>)
+#>     }
+#>     {?work cdm:resource_legal_is_about_concept_directory-code ?value.
+#>     }
+#>     UNION
+#>     {?work cdm:resource_legal_is_about_concept_directory-code ?directory.
+#>       ?value skos:narrower+ ?directory.
+#>     }
+#>     
+#>     ?work cdm:resource_legal_id_sector ?sector.
+#>     FILTER(str(?sector)='3')
+#>      
+#>  FILTER not exists{?work cdm:work_has_resource-type <http://publications.europa.eu/resource/authority/resource-type/CORRIGENDUM>} OPTIONAL{?work cdm:resource_legal_id_celex ?celex.} }

Now that we have a query, we are ready to run it.

elx_run_query(): Execute SPARQL queries

elx_run_query() sends SPARQL queries to a pre-specified endpoint. The function takes the query string as the main argument, which means you can manually pass it any working SPARQL query (relevant to official EU publications).

-
results <- elx_run_query(query = query_dir)
-
-# the functions are compatible with piping
-# 
-# elx_make_query("directive") %>% 
-#   elx_run_query()
-
as_tibble(results)
-#> # A tibble: 4,192 x 3
-#>   work                                   type                            celex  
-#>   <chr>                                  <chr>                           <chr>  
-#> 1 http://publications.europa.eu/resourc~ http://publications.europa.eu/~ 31979L~
-#> 2 http://publications.europa.eu/resourc~ http://publications.europa.eu/~ 31989L~
-#> 3 http://publications.europa.eu/resourc~ http://publications.europa.eu/~ 31984L~
-#> 4 http://publications.europa.eu/resourc~ http://publications.europa.eu/~ 31966L~
-#> # ... with 4,188 more rows
+
results <- elx_run_query(query = query_dir)
+
+# the functions are compatible with piping
+# 
+# elx_make_query("directive") %>% 
+#   elx_run_query()
+
as_tibble(results)
+#> # A tibble: 4,317 x 3
+#>   work                                   type                            celex  
+#>   <chr>                                  <chr>                           <chr>  
+#> 1 http://publications.europa.eu/resourc~ http://publications.europa.eu/~ 31979L~
+#> 2 http://publications.europa.eu/resourc~ http://publications.europa.eu/~ 31989L~
+#> 3 http://publications.europa.eu/resourc~ http://publications.europa.eu/~ 31984L~
+#> 4 http://publications.europa.eu/resourc~ http://publications.europa.eu/~ 31966L~
+#> # ... with 4,313 more rows

The function outputs a data.frame where each column corresponds to one of the requested variables, while the rows accumulate observations of the resource type satisfying the query criteria. Obviously, the more data is to be returned, the longer the execution time, varying from a few seconds to several minutes, depending also on your connection.

The first column always contains the unique URI of a “work” (legislative act or court judgment) which identifies each resource in Cellar. Several human-readable identifiers are normally associated with each “work” but the most useful one is CELEX, retrieved by default.2

One column you should always pay attention to is type (as in resource_type). The URIs contained there reflect the FILTER argument in the SPARQL query, which is manually pre-specified. All resources are indexed as being of one type or another. For example, when retrieving directives, the results are going to return also delegated directives, which might not be desirable, depending on your needs. You can filter results by type to make the necessary adjustments. The queries are expansive by default in the spirit of erring on the side of over-inclusiveness rather than vice versa.

-
head(results$type,5)
-#> [1] "http://publications.europa.eu/resource/authority/resource-type/DIR"
-#> [2] "http://publications.europa.eu/resource/authority/resource-type/DIR"
-#> [3] "http://publications.europa.eu/resource/authority/resource-type/DIR"
-#> [4] "http://publications.europa.eu/resource/authority/resource-type/DIR"
-#> [5] "http://publications.europa.eu/resource/authority/resource-type/DIR"
-
-results %>% 
-  distinct(type)
-#> # A tibble: 3 x 1
-#>   type                                                                   
-#>   <chr>                                                                  
-#> 1 http://publications.europa.eu/resource/authority/resource-type/DIR     
-#> 2 http://publications.europa.eu/resource/authority/resource-type/DIR_IMPL
-#> 3 http://publications.europa.eu/resource/authority/resource-type/DIR_DEL
+
head(results$type,5)
+#> [1] "http://publications.europa.eu/resource/authority/resource-type/DIR"
+#> [2] "http://publications.europa.eu/resource/authority/resource-type/DIR"
+#> [3] "http://publications.europa.eu/resource/authority/resource-type/DIR"
+#> [4] "http://publications.europa.eu/resource/authority/resource-type/DIR"
+#> [5] "http://publications.europa.eu/resource/authority/resource-type/DIR"
+
+results %>% 
+  distinct(type)
+#> # A tibble: 3 x 1
+#>   type                                                                   
+#>   <chr>                                                                  
+#> 1 http://publications.europa.eu/resource/authority/resource-type/DIR     
+#> 2 http://publications.europa.eu/resource/authority/resource-type/DIR_IMPL
+#> 3 http://publications.europa.eu/resource/authority/resource-type/DIR_DEL

The data is returned in the long format, which means that rows are recycled up to the length of the variable with the most data points. For example, if 20 directives are returned, each with two legal bases, the resulting data.frame will have 40 rows. Some variables, such as dates, contain unexpectedly several entries for some documents. You should always check the number of unique identifiers in the results instead of assuming that each row is a unique observation.

EuroVoc descriptors

EuroVoc is a multilingual thesaurus, keywords from which are used to describe the content of European Union documents. Most resource types that can be retrieved with the pre-defined queries in this package can be accompanied by EuroVoc keywords and these can be retrieved as other variables.

- -

By default, the endpoint returns the EuroVoc concept codes rather than the labels (keywords). The function elx_label_eurovoc() needs to be called to obtain a look-up table with the labels.

+rec_eurovoc <- elx_make_query("recommendation", include_eurovoc = TRUE, limit = 10) %>% + elx_run_query() # truncated results for sake of the example + +rec_eurovoc %>% + select(celex, eurovoc) +#> # A tibble: 10 x 2 +#> celex eurovoc +#> <chr> <chr> +#> 1 32012H0090 http://eurovoc.europa.eu/1425 +#> 2 31962H0816 http://eurovoc.europa.eu/1004 +#> 3 31974H0435 http://eurovoc.europa.eu/1085 +#> 4 31996H0592 http://eurovoc.europa.eu/1076 +#> # ... with 6 more rows
+

By default, the endpoint returns the EuroVoc concept codes rather than the labels (keywords). The function elx_label_eurovoc() needs to be called to obtain a look-up table with the labels.

+
eurovoc_lookup <- elx_label_eurovoc(uri_eurovoc = rec_eurovoc$eurovoc)
+
+print(eurovoc_lookup)
+#> # A tibble: 9 x 2
+#>   eurovoc                       labels         
+#>   <chr>                         <chr>          
+#> 1 http://eurovoc.europa.eu/1085 France         
+#> 2 http://eurovoc.europa.eu/1442 food inspection
+#> 3 http://eurovoc.europa.eu/1076 form           
+#> 4 http://eurovoc.europa.eu/1318 Germany        
+#> # ... with 5 more rows

The results include labels only for unique identifiers, but with dplyr::left_join() it is straightforward to append the labels to the entire dataset.

-

-rec_eurovoc %>% 
-  left_join(eurovoc_lookup)
-#> Joining, by = "eurovoc"
-#> # A tibble: 10 x 5
-#>   work                      type                   celex    eurovoc      labels 
-#>   <chr>                     <chr>                  <chr>    <chr>        <chr>  
-#> 1 http://publications.euro~ http://publications.e~ 31962H0~ http://euro~ welfare
-#> 2 http://publications.euro~ http://publications.e~ 32015H0~ http://euro~ tax sy~
-#> 3 http://publications.euro~ http://publications.e~ 32016H0~ http://euro~ tax sy~
-#> 4 http://publications.euro~ http://publications.e~ 32017H0~ http://euro~ tax sy~
-#> # ... with 6 more rows
+
rec_eurovoc %>% 
+  left_join(eurovoc_lookup)
+#> Joining, by = "eurovoc"
+#> # A tibble: 10 x 5
+#>   work                      type                   celex  eurovoc      labels   
+#>   <chr>                     <chr>                  <chr>  <chr>        <chr>    
+#> 1 http://publications.euro~ http://publications.e~ 32012~ http://euro~ consumer~
+#> 2 http://publications.euro~ http://publications.e~ 31962~ http://euro~ welfare  
+#> 3 http://publications.euro~ http://publications.e~ 31974~ http://euro~ France   
+#> 4 http://publications.euro~ http://publications.e~ 31996~ http://euro~ form     
+#> # ... with 6 more rows

As elsewhere in the API, we can tap into the multilingual nature of EU documents also when it comes to the EuroVoc keywords. Moreover, most concepts in the thesaurus are associated with alternative labels; these can be returned as well (separated by a comma).

-
eurovoc_lookup <- elx_label_eurovoc(uri_eurovoc = rec_eurovoc$eurovoc,
-                                    alt_labels = TRUE,
-                                    language = "sk")
-
-rec_eurovoc %>% 
-  left_join(eurovoc_lookup) %>% 
-  select(celex, eurovoc, labels)
-#> Joining, by = "eurovoc"
-#> # A tibble: 10 x 3
-#>   celex          eurovoc                       labels                      
-#>   <chr>          <chr>                         <chr>                       
-#> 1 31962H0816     http://eurovoc.europa.eu/1004 blahobyt                    
-#> 2 32015H0818(10) http://eurovoc.europa.eu/1021 danová sústava,danový systém
-#> 3 32016H0818(09) http://eurovoc.europa.eu/1021 danová sústava,danový systém
-#> 4 32017H0809(10) http://eurovoc.europa.eu/1021 danová sústava,danový systém
-#> # ... with 6 more rows
+
eurovoc_lookup <- elx_label_eurovoc(uri_eurovoc = rec_eurovoc$eurovoc,
+                                    alt_labels = TRUE,
+                                    language = "sk")
+
+rec_eurovoc %>% 
+  left_join(eurovoc_lookup) %>% 
+  select(celex, eurovoc, labels)
+#> Joining, by = "eurovoc"
+#> # A tibble: 10 x 3
+#>   celex     eurovoc                   labels                                    
+#>   <chr>     <chr>                     <chr>                                     
+#> 1 32012H00~ http://eurovoc.europa.eu~ informácie pre spotrebitela,vzdelávanie s~
+#> 2 31962H08~ http://eurovoc.europa.eu~ blahobyt                                  
+#> 3 31974H04~ http://eurovoc.europa.eu~ Francúzska republika,Francúzsko           
+#> 4 31996H05~ http://eurovoc.europa.eu~ formulár                                  
+#> # ... with 6 more rows

elx_fetch_data(): Fire GET requests

A core contribution of the SPARQL requests is that we obtain a comprehensive list of identifiers that we can subsequently use to obtain more data relating to the document in question. While the results of the SPARQL queries are useful also for webscraping (with the rvest package), the function elx_fetch_data() enables us to fire GET requests to retrieve data on documents with known identifiers (including Cellar URI).

One of the most sought-after data in the Eur-Lex dataverse is the text. It is possible now to automate the pipeline for downloading html and plain texts from Eur-Lex. Similarly, you can retrieve the title of the document. For both you can specify also the desired language (English by default). Other metadata might be added in the future.

-
# the function is not vectorized by default
-elx_fetch_data(results$work[1],"title")
-#> [1] "Council Directive 79/173/EEC of 6 February 1979 on the programme for the acceleration and guidance of collective irrigation works in Corsica"
-
-# we can use purrr::map() to play that role
-library(purrr)
-
-dir_titles <- results[1:10,] %>% # take the first 10 directives only to save time
-  mutate(title = map_chr(work,elx_fetch_data, "title")) %>% 
-  as_tibble() %>% 
-  select(celex, title)
-
-print(dir_titles)
-#> # A tibble: 10 x 2
-#>   celex      title                                                              
-#>   <chr>      <chr>                                                              
-#> 1 31979L0173 Council Directive 79/173/EEC of 6 February 1979 on the programme f~
-#> 2 31989L0194 Council Directive 89/194/EEC of 13 March 1989 amending Directive 6~
-#> 3 31984L0378 Council Directive 84/378/EEC of 28 June 1984 amending the Annexes ~
-#> 4 31966L0683 Commission Directive 66/683/EEC of 7 November 1966 eliminating all~
-#> # ... with 6 more rows
-

Note that text requests are by far the most time-intensive; requesting the full text for thousands of documents is liable to extend the run-time into hours. Currently, no method for downloading text in non-html/plain formats is implemented, which means pdf-only texts will be missing from the results.3

+
# the function is not vectorized by default
+elx_fetch_data(results$work[1],"title")
+#> [1] "Council Directive 79/173/EEC of 6 February 1979 on the programme for the acceleration and guidance of collective irrigation works in Corsica"
+
+# we can use purrr::map() to play that role
+library(purrr)
+
+dir_titles <- results[1:10,] %>% # take the first 10 directives only to save time
+  mutate(title = map_chr(work,elx_fetch_data, "title")) %>% 
+  as_tibble() %>% 
+  select(celex, title)
+
+print(dir_titles)
+#> # A tibble: 10 x 2
+#>   celex      title                                                              
+#>   <chr>      <chr>                                                              
+#> 1 31979L0173 Council Directive 79/173/EEC of 6 February 1979 on the programme f~
+#> 2 31989L0194 Council Directive 89/194/EEC of 13 March 1989 amending Directive 6~
+#> 3 31984L0378 Council Directive 84/378/EEC of 28 June 1984 amending the Annexes ~
+#> 4 31966L0683 Commission Directive 66/683/EEC of 7 November 1966 eliminating all~
+#> # ... with 6 more rows
+

Note that text requests are by far the most time-intensive; requesting the full text for thousands of documents is liable to extend the run-time into hours. Texts are retrieved from html by priority, but methods for pdfs and .docs are also implemented.3 The function even handles multi-document resources (by pasting them together).

Application

In this section I showcase a simple application of eurlex on making overviews of EU legislation. First, we collate data on directives.

-
dirs <- elx_make_query(resource_type = "directive", include_date = TRUE, include_force = TRUE) %>% 
-  elx_run_query() %>% 
-  rename(date = `callret-3`)
+
dirs <- elx_make_query(resource_type = "directive", include_date = TRUE, include_force = TRUE) %>% 
+  elx_run_query() %>% 
+  rename(date = `callret-3`)

Let’s calculate the proportion of directives currently in force in the entire set of directives ever adopted. This variable offers a particularly good demonstration of the usefulness of the package to retrieve EU law data, because it changes every day, as new acts enter into force and old ones drop out. Regularly scraping webpages for this purpose and scale is simply impractical and disproportional.

-
library(ggplot2)
-
-dirs %>% 
-  count(force) %>% 
-  ggplot(aes(x = force, y = n)) +
-  geom_col()
-

+
library(ggplot2)
+
+dirs %>% 
+  count(force) %>% 
+  ggplot(aes(x = force, y = n)) +
+  geom_col()
+

Directives become naturally outdated with time. It might be all the more interesting to see which older acts are thus still surviving.

-
dirs %>% 
-  ggplot(aes(x = as.Date(date), y = celex)) +
-  geom_point(aes(color = force), alpha = 0.1) +
-  theme(axis.text.y = element_blank(),
-        axis.line.y = element_blank(),
-        axis.ticks.y = element_blank())
-

+
dirs %>% 
+  filter(!is.na(force)) %>% 
+  mutate(date = as.Date(date)) %>% 
+  ggplot(aes(x = date, y = celex)) +
+  geom_point(aes(color = force), alpha = 0.1) +
+  theme(axis.text.y = element_blank(),
+        axis.line.y = element_blank(),
+        axis.ticks.y = element_blank())
+

We want to know a bit more about the directives from 1970s that are still in force today. Their titles could give us a clue.

-
dirs_1970_title <- dirs %>% 
-  filter(between(as.Date(date), as.Date("1970-01-01"), as.Date("1980-01-01")),
-         force == "true") %>% 
-  mutate(title = map_chr(work,elx_fetch_data,"title")) %>% 
-  as_tibble()
-
-print(dirs_1970_title)
-#> # A tibble: 78 x 6
-#>   work                 type               celex  date  force title              
-#>   <chr>                <chr>              <chr>  <chr> <chr> <chr>              
-#> 1 http://publications~ http://publicatio~ 31975~ 1975~ true  Council Directive ~
-#> 2 http://publications~ http://publicatio~ 31977~ 1977~ true  First Commission D~
-#> 3 http://publications~ http://publicatio~ 31977~ 1977~ true  Council Directive ~
-#> 4 http://publications~ http://publicatio~ 31973~ 1973~ true  Council Directive ~
-#> # ... with 74 more rows
+
dirs_1970_title <- dirs %>% 
+  filter(between(as.Date(date), as.Date("1970-01-01"), as.Date("1980-01-01")),
+         force == "true") %>% 
+  mutate(title = map_chr(work,elx_fetch_data,"title")) %>% 
+  as_tibble()
+
+print(dirs_1970_title)
+#> # A tibble: 70 x 6
+#>   work                 type               celex  date  force title              
+#>   <chr>                <chr>              <chr>  <chr> <chr> <chr>              
+#> 1 http://publications~ http://publicatio~ 31975~ 1975~ true  Council Directive ~
+#> 2 http://publications~ http://publicatio~ 31977~ 1977~ true  First Commission D~
+#> 3 http://publications~ http://publicatio~ 31977~ 1977~ true  Council Directive ~
+#> 4 http://publications~ http://publicatio~ 31973~ 1973~ true  Council Directive ~
+#> # ... with 66 more rows

I will use the tidytext package to get a quick idea of what the legislation is about.

-
library(tidytext)
-library(wordcloud)
-
-dirs_1970_title %>% 
-  select(celex,title) %>% 
-  unnest_tokens(word, title) %>% 
-  count(celex, word, sort = TRUE) %>% 
-  filter(!grepl("\\d", word)) %>% 
-  bind_tf_idf(word, celex, n) %>% 
-  with(wordcloud(word, tf_idf, max.words = 40, scale = c(1.8,0.1)))
-

+
library(tidytext)
+library(wordcloud)
+
+dirs_1970_title %>% 
+  select(celex,title) %>% 
+  unnest_tokens(word, title) %>% 
+  count(celex, word, sort = TRUE) %>% 
+  filter(!grepl("\\d", word)) %>% 
+  bind_tf_idf(word, celex, n) %>% 
+  with(wordcloud(word, tf_idf, max.words = 40, scale = c(1.8,0.1)))
+

I use term-frequency inverse-document frequency (tf-idf) to weight the importance of the words in the wordcloud. If we used pure frequencies, the wordcloud would largely consist of words conveying little meaning (“the”, “and”, …).

-

This is an extremely basic application of the eurlex package. Much more sophisticated methods can be used to analyse both the content and metadata of European Union legislation. If the package is useful for your research, please consider citing it.

+

This is an extremely basic application of the eurlex package. Much more sophisticated methods can be used to analyse both the content and metadata of European Union legislation. If the package is useful for your research, please consider citing the accompanying paper.4

@@ -588,6 +629,7 @@

Application

  • Note, however, that not all resource types will work properly with the pre-specified query.↩

  • Occasionally, you may encounter legal acts without CELEX numbers, especially when digging through older legislation. It is good to report these to the Eur-Lex helpdesk.↩

  • It is worth pointing out that the html and pdf contents of older case law differs. Whereas typically the html file is only going to contain a summary and grounds of a judgment, the pdf should also contain background to the dispute.↩

    • +

  • Michal Ovádek (2021) Facilitating access to data on European Union laws, Political Research Exchange, 3:1, DOI: 10.1080/2474736X.2020.1870150↩

    • diff --git a/docs/404.html b/docs/404.html index 45cec7e..21866c3 100644 --- a/docs/404.html +++ b/docs/404.html @@ -79,7 +79,7 @@ eurlex - 0.3.4 + 0.3.5 diff --git a/docs/articles/eurlexpkg.html b/docs/articles/eurlexpkg.html index 5739615..dfc9641 100644 --- a/docs/articles/eurlexpkg.html +++ b/docs/articles/eurlexpkg.html @@ -39,7 +39,7 @@ eurlex - 0.3.4 + 0.3.5 @@ -105,6 +105,7 @@

    The eurlex package

    The eurlex package currently envisions the typical use-case to consist of getting bulk information about EU legislation into R as fast as possible. The package contains three core functions to achieve that objective: elx_make_query() to create pre-defined or customized SPARQL queries; elx_run_query() to execute the pre-made or any other manually input query; and elx_fetch_data() to fire GET requests for certain metadata to the REST API.

    +

    The package also contains largely self-explanatory functions for retrieving data on EU court cases (elx_curia_list()) and Council votes (elx_council_votes()) from outside Eur-Lex.

    elx_make_query(): Generate SPARQL queries

    @@ -116,7 +117,7 @@

    Currently, it is possible to choose from among a host of resource types, including directives, regulations and even case law (see function description for the full list). It is also possible to manually specify a resource type from the eligible list.1

    The choice of resource type is then reflected in the SPARQL query generated by the function:

    query_dir %>%
-  glue::as_glue() # for nicer printing
+  cat() # for nicer printing
 #> PREFIX cdm: <http://publications.europa.eu/ontology/cdm#>
 #>   PREFIX annot: <http://publications.europa.eu/ontology/annotation#>
 #>   PREFIX skos:<http://www.w3.org/2004/02/skos/core#>
@@ -130,7 +131,7 @@ 
    
 #>  FILTER not exists{?work cdm:work_has_resource-type <http://publications.europa.eu/resource/authority/resource-type/CORRIGENDUM>} OPTIONAL{?work cdm:resource_legal_id_celex ?celex.} }
 
 elx_make_query(resource_type = "caselaw") %>%
-  glue::as_glue()
+  cat()
 #> PREFIX cdm: <http://publications.europa.eu/ontology/cdm#>
 #>   PREFIX annot: <http://publications.europa.eu/ontology/annotation#>
 #>   PREFIX skos:<http://www.w3.org/2004/02/skos/core#>
@@ -150,7 +151,7 @@ 
    
 #>   ?type=<http://publications.europa.eu/resource/authority/resource-type/OPIN_AG>) OPTIONAL{?work cdm:resource_legal_id_celex ?celex.} }
 
 elx_make_query(resource_type = "manual", manual_type = "SWD") %>%
-  glue::as_glue()
+  cat()
 #> PREFIX cdm: <http://publications.europa.eu/ontology/cdm#>
 #>   PREFIX annot: <http://publications.europa.eu/ontology/annotation#>
 #>   PREFIX skos:<http://www.w3.org/2004/02/skos/core#>
@@ -162,9 +163,9 @@ 
    
 #>  FILTER not exists{?work cdm:work_has_resource-type <http://publications.europa.eu/resource/authority/resource-type/CORRIGENDUM>} OPTIONAL{?work cdm:resource_legal_id_celex ?celex.} }

    There are various ways of querying the same information in the Cellar database due to the existence of several overlapping classes and identifiers describing the same resources. The queries generated by the function should offer a reliable way of obtaining exhaustive results, as they have been validated by the helpdesk of the Publication Office. At the same time, it is always possible there will be issues either on the query or the database side; please report any you encounter through Github.

    The other arguments in elx_make_query() relate to additional metadata to be returned. The results include by default the CELEX number and exclude corrigenda (corrections of errors in legislation). Other data needs to be opted into. Make sure to select ones that are logically compatible (e.g. case law does not have a legal basis). More options should be added in the future.

    -

    Note that availability of data for each variable has an impact on the results. The data frame returned by the query will be shrunken to the size of the variable with most missing data. It is recommended to always compare results from a desired query to a minimal query requesting only celex ids.

    +

    Note that availability of data for each variable might have an impact on the results. The data frame returned by the query might be shrunken to the size of the variable with most missing data. It is recommended to always compare results from a desired query to a minimal query requesting only celex ids.

    elx_make_query(resource_type = "directive", include_date = TRUE, include_force = TRUE) %>%
-  glue::as_glue()
+  cat()
 #> PREFIX cdm: <http://publications.europa.eu/ontology/cdm#>
 #>   PREFIX annot: <http://publications.europa.eu/ontology/annotation#>
 #>   PREFIX skos:<http://www.w3.org/2004/02/skos/core#>
@@ -180,7 +181,7 @@ 
    
 # minimal query: elx_make_query(resource_type = "directive")
 
 elx_make_query(resource_type = "recommendation", include_date = TRUE, include_lbs = TRUE) %>%
-  glue::as_glue()
+  cat()
 #> PREFIX cdm: <http://publications.europa.eu/ontology/cdm#>
 #>   PREFIX annot: <http://publications.europa.eu/ontology/annotation#>
 #>   PREFIX skos:<http://www.w3.org/2004/02/skos/core#>
@@ -204,31 +205,62 @@ 
    
 #>     ?bn annot:comment_on_legal_basis ?lbsuffix}} }
 
 # minimal query: elx_make_query(resource_type = "recommendation")
    +

    You can also decide to not specify any resource types, in which case all types of documents will be returned. As there are over a million documents with a CELEX identifier, this is likely not efficient for a majority of users. But since version 0.3.5 it is possible to request documents belonging to a particular “sector” or directory code.

    +
    # request documents from directory 18 ("Common Foreign and Security Policy")
+# and sector 3 ("Legal acts")
+
+elx_make_query(resource_type = "any",
+               directory = "18",
+               sector = 3) %>%
+  cat()
+#> PREFIX cdm: <http://publications.europa.eu/ontology/cdm#>
+#>   PREFIX annot: <http://publications.europa.eu/ontology/annotation#>
+#>   PREFIX skos:<http://www.w3.org/2004/02/skos/core#>
+#>   PREFIX dc:<http://purl.org/dc/elements/1.1/>
+#>   PREFIX xsd:<http://www.w3.org/2001/XMLSchema#>
+#>   PREFIX rdf:<http://www.w3.org/1999/02/22-rdf-syntax-ns#>
+#>   PREFIX owl:<http://www.w3.org/2002/07/owl#>
+#>   select distinct ?work ?type ?celex where{
+#>     VALUES (?value)
+#>     { (<http://publications.europa.eu/resource/authority/fd_555/18>)
+#>       (<http://publications.europa.eu/resource/authority/dir-eu-legal-act/18>)
+#>     }
+#>     {?work cdm:resource_legal_is_about_concept_directory-code ?value.
+#>     }
+#>     UNION
+#>     {?work cdm:resource_legal_is_about_concept_directory-code ?directory.
+#>       ?value skos:narrower+ ?directory.
+#>     }
+#>     
+#>     ?work cdm:resource_legal_id_sector ?sector.
+#>     FILTER(str(?sector)='3')
+#>      
+#>  FILTER not exists{?work cdm:work_has_resource-type <http://publications.europa.eu/resource/authority/resource-type/CORRIGENDUM>} OPTIONAL{?work cdm:resource_legal_id_celex ?celex.} }

    Now that we have a query, we are ready to run it.

    elx_run_query(): Execute SPARQL queries

    elx_run_query() sends SPARQL queries to a pre-specified endpoint. The function takes the query string as the main argument, which means you can manually pass it any working SPARQL query (relevant to official EU publications).

    -
    results <- elx_run_query(query = query_dir)
+
    results <- elx_run_query(query = query_dir)
 
 # the functions are compatible with piping
 # 
 # elx_make_query("directive") %>% 
 #   elx_run_query()
    
-
    as_tibble(results)
-#> # A tibble: 4,316 x 3
+
    as_tibble(results)
+#> # A tibble: 4,317 x 3
 #>   work                                   type                            celex  
 #>   <chr>                                  <chr>                           <chr>  
 #> 1 http://publications.europa.eu/resourc~ http://publications.europa.eu/~ 31979L~
 #> 2 http://publications.europa.eu/resourc~ http://publications.europa.eu/~ 31989L~
 #> 3 http://publications.europa.eu/resourc~ http://publications.europa.eu/~ 31984L~
 #> 4 http://publications.europa.eu/resourc~ http://publications.europa.eu/~ 31966L~
-#> # ... with 4,312 more rows
    
+#> # ... with 4,313 more rows
    
 
    The function outputs a data.frame where each column corresponds to one of the requested variables, while the rows accumulate observations of the resource type satisfying the query criteria. Obviously, the more data is to be returned, the longer the execution time, varying from a few seconds to several minutes, depending also on your connection.
    
 
    The first column always contains the unique URI of a “work” (legislative act or court judgment) which identifies each resource in Cellar. Several human-readable identifiers are normally associated with each “work” but the most useful one is CELEX, retrieved by default.2
    
 
    One column you should always pay attention to is type (as in resource_type). The URIs contained there reflect the FILTER argument in the SPARQL query, which is manually pre-specified. All resources are indexed as being of one type or another. For example, when retrieving directives, the results are going to return also delegated directives, which might not be desirable, depending on your needs. You can filter results by type to make the necessary adjustments. The queries are expansive by default in the spirit of erring on the side of over-inclusiveness rather than vice versa.
    
-
    head(results$type,5)
+
    head(results$type,5)
 #> [1] "http://publications.europa.eu/resource/authority/resource-type/DIR"
 #> [2] "http://publications.europa.eu/resource/authority/resource-type/DIR"
 #> [3] "http://publications.europa.eu/resource/authority/resource-type/DIR"
@@ -248,7 +280,7 @@ 
    
 
    
 EuroVoc descriptors
    
 
    EuroVoc is a multilingual thesaurus, keywords from which are used to describe the content of European Union documents. Most resource types that can be retrieved with the pre-defined queries in this package can be accompanied by EuroVoc keywords and these can be retrieved as other variables.
    
-
    +
     rec_eurovoc <- elx_make_query("recommendation", include_eurovoc = TRUE, limit = 10) %>%
   elx_run_query() # truncated results for sake of the example
 
@@ -263,7 +295,7 @@ 
    
 #> 4 31996H0592 http://eurovoc.europa.eu/1076
 #> # ... with 6 more rows
    
 
    By default, the endpoint returns the EuroVoc concept codes rather than the labels (keywords). The function elx_label_eurovoc() needs to be called to obtain a look-up table with the labels.
    
-
    eurovoc_lookup <- elx_label_eurovoc(uri_eurovoc = rec_eurovoc$eurovoc)
+
    eurovoc_lookup <- elx_label_eurovoc(uri_eurovoc = rec_eurovoc$eurovoc)
 
 print(eurovoc_lookup)
 #> # A tibble: 9 x 2
@@ -275,7 +307,7 @@ 
    
 #> 4 http://eurovoc.europa.eu/1318 Germany        
 #> # ... with 5 more rows
    
 
    The results include labels only for unique identifiers, but with dplyr::left_join() it is straightforward to append the labels to the entire dataset.
    
-
    rec_eurovoc %>%
+
    rec_eurovoc %>%
   left_join(eurovoc_lookup)
 #> Joining, by = "eurovoc"
 #> # A tibble: 10 x 5
@@ -287,7 +319,7 @@ 
    
 #> 4 http://publications.euro~ http://publications.e~ 31996~ http://euro~ form     
 #> # ... with 6 more rows
    
 
    As elsewhere in the API, we can tap into the multilingual nature of EU documents also when it comes to the EuroVoc keywords. Moreover, most concepts in the thesaurus are associated with alternative labels; these can be returned as well (separated by a comma).
    
-
    eurovoc_lookup <- elx_label_eurovoc(uri_eurovoc = rec_eurovoc$eurovoc,
+
    eurovoc_lookup <- elx_label_eurovoc(uri_eurovoc = rec_eurovoc$eurovoc,
                                     alt_labels = TRUE,
                                     language = "sk")
 
@@ -310,7 +342,7 @@ 
    
 elx_fetch_data(): Fire GET requests
    
 
    A core contribution of the SPARQL requests is that we obtain a comprehensive list of identifiers that we can subsequently use to obtain more data relating to the document in question. While the results of the SPARQL queries are useful also for webscraping (with the rvest package), the function elx_fetch_data() enables us to fire GET requests to retrieve data on documents with known identifiers (including Cellar URI).
    
 
    One of the most sought-after data in the Eur-Lex dataverse is the text. It is possible now to automate the pipeline for downloading html and plain texts from Eur-Lex. Similarly, you can retrieve the title of the document. For both you can specify also the desired language (English by default). Other metadata might be added in the future.
    
-
    # the function is not vectorized by default
+
    # the function is not vectorized by default
 elx_fetch_data(results$work[1],"title")
 #> [1] "Council Directive 79/173/EEC of 6 February 1979 on the programme for the acceleration and guidance of collective irrigation works in Corsica"
 
@@ -338,11 +370,11 @@ 
    
 
    
 Application
    
 
    In this section I showcase a simple application of eurlex on making overviews of EU legislation. First, we collate data on directives.
    
-
    dirs <- elx_make_query(resource_type = "directive", include_date = TRUE, include_force = TRUE) %>%
+
    dirs <- elx_make_query(resource_type = "directive", include_date = TRUE, include_force = TRUE) %>%
   elx_run_query() %>%
   rename(date = `callret-3`)
    
 
    Let’s calculate the proportion of directives currently in force in the entire set of directives ever adopted. This variable offers a particularly good demonstration of the usefulness of the package to retrieve EU law data, because it changes every day, as new acts enter into force and old ones drop out. Regularly scraping webpages for this purpose and scale is simply impractical and disproportional.
    
-
    library(ggplot2)
+
    library(ggplot2)
 
 dirs %>%
   count(force) %>%
@@ -350,15 +382,17 @@ 
    
   geom_col()
    
 
    
 
    Directives become naturally outdated with time. It might be all the more interesting to see which older acts are thus still surviving.
    
-
    dirs %>%
-  ggplot(aes(x = as.Date(date), y = celex)) +
+
    dirs %>%
+  filter(!is.na(force)) %>%
+  mutate(date = as.Date(date)) %>%
+  ggplot(aes(x = date, y = celex)) +
   geom_point(aes(color = force), alpha = 0.1) +
   theme(axis.text.y = element_blank(),
         axis.line.y = element_blank(),
         axis.ticks.y = element_blank())
    
-
    
+
    
 
    We want to know a bit more about the directives from 1970s that are still in force today. Their titles could give us a clue.
    
-
    dirs_1970_title <- dirs %>%
+
    dirs_1970_title <- dirs %>%
   filter(between(as.Date(date), as.Date("1970-01-01"), as.Date("1980-01-01")),
          force == "true") %>%
   mutate(title = map_chr(work,elx_fetch_data,"title")) %>%
@@ -374,7 +408,7 @@ 
    
 #> 4 http://publications~ http://publicatio~ 31973~ 1973~ true  Council Directive ~
 #> # ... with 66 more rows
    
 
    I will use the tidytext package to get a quick idea of what the legislation is about.
    
-
    library(tidytext)
+
    library(tidytext)
 library(wordcloud)
 
 dirs_1970_title %>%
@@ -386,7 +420,7 @@ 
    
   with(wordcloud(word, tf_idf, max.words = 40, scale = c(1.8,0.1)))
    
 
    
 
    I use term-frequency inverse-document frequency (tf-idf) to weight the importance of the words in the wordcloud. If we used pure frequencies, the wordcloud would largely consist of words conveying little meaning (“the”, “and”, …).
    
-
    This is an extremely basic application of the eurlex package. Much more sophisticated methods can be used to analyse both the content and metadata of European Union legislation. If the package is useful for your research, please consider citing it.
    
+
    This is an extremely basic application of the eurlex package. Much more sophisticated methods can be used to analyse both the content and metadata of European Union legislation. If the package is useful for your research, please consider citing the accompanying paper.4
    
 
    
 
    
 
    
@@ -394,6 +428,7 @@ 
    
 
  • Note, however, that not all resource types will work properly with the pre-specified query.↩
    • 
 
  • Occasionally, you may encounter legal acts without CELEX numbers, especially when digging through older legislation. It is good to report these to the Eur-Lex helpdesk.↩
    • 
 
  • It is worth pointing out that the html and pdf contents of older case law differs. Whereas typically the html file is only going to contain a summary and grounds of a judgment, the pdf should also contain background to the dispute.↩
    • 
+
  • Michal Ovádek (2021) Facilitating access to data on European Union laws, Political Research Exchange, 3:1, DOI: 10.1080/2474736X.2020.1870150↩
    • 
 
 
    
   
    
diff --git a/docs/articles/eurlexpkg_files/figure-html/unnamed-chunk-9-1.png b/docs/articles/eurlexpkg_files/figure-html/unnamed-chunk-9-1.png
index 8f93a1014e68a600b0a417adbd54a19cff0423d4..e7a0faf45fbe53d8ce055cb56bc98cab8c43d50a 100644
GIT binary patch
literal 56081
zcma%jby$?!7xh?h6%>(_5FP1|4#}$yT{4mb(%s!*T)L#CB?hE(=u*0+hVB@8=#KAY
z#^3k-`~7jB`#xUJJLl}P_S$Q&bKV>HUP1EV1BwSA5a^+_6a)$a-E#$j?&$q<8~Bq!
zUC4dlU-xaLG#o%6YX6&mZiO%J8i7F1LDG;n%C1S9(=O2~YF?;QbbaH=QkTldksQIF
zBTr;_-ro8KGH9|R#lXwKZr2p}
z{L8)D|Gi&THO+A-T<@9>Y;;6VizVPpzI!jrCZaNe9@PoY9Z^X=H}N|2XjFTRx(mz;
z1e$aIbNLAK1gdxxgoAmqoqo@QdHA&9mwpTLq{UDC5c4oslV*x}0F6Erzk_+IvSy^f
zJTy7I8^HdzLdyTISM8otWBz`{mt$cc&sW_P4UO8wvdM*qb?40RrW&h;M(Ztor4UVz
z9OZS5T5aMo4N+=Ia3oB~I^?7wAmpqez%l<)jJ{pmP?UX}r#fX~X4c`>Xb^VJ+Ck$y
ziT(D3PYJ&Ic${r&#j-86zSYjp)?noNF|0N*P2J$a)4Tnu6}<2|j?RoJ%s7NCsdi5u
zT%Z=3%w?>RlZ)IJw_D!WcjJtuH@E$vlYK{<3%jZ+V}DI*;JtSUPv?4;`-}1erCvk^
zhtz20{V?f?bjnDnWZ7w)k+lx|?Nq>`5i6^F=s3C_Y*z8|fk8o36MDt1bAJNaXLu?`
z`BUeVFsdPYJybM>(A5kZ#<>Y9Yp1s6xoU;u0b^N+w~KQj%R(R{;mbie`Nz0)8^RIX
z8r~ca#0}}${e(pd52Y@^ug%{JwMRXYE1nFMYANnS_w_`;@E8&
z)#d9Y-M?sd*rzr)ijz@v5wrlqT7t+U-nU46mX0zejuTG+!wg;KQhWBM`(l}6B0J2L
zhCOXesOtZ0-IIBEeic*rYq|B@V!4DTp5m!-?5#vHEC#f`vKR7-32;xA@V#G`i%eRl
z1M7?5j$om2@5+A=VaMH-kc9PyGMH29=2fKOGOH(vluShI<
z)spKqss5N6p{Bh-#V7Q(ZQlL|b}F4-dmR?VoRkvz2q{{UKuCVTlU)cj0v34y*|r+?
zN5D4WMd8h#2L`f*l2Z_s(cP75%+0^}uclpi{%LF0<$7@L!87Ns_ZJn>;l1pawcI9g
zp@-srgq}Ss)6beBfs&_QUYAT5|F_E)Wm4PGA)5}YUcBQc7i%+92jo%Kfz+=
z{+>dyY6
zAysf{U3kK>fUI>}K$|T{Oa~^4l!deq4S=fy7#1KYPQjwzjeoAcD$8ky{A7s80SB;<~(1x{!c|n&dZua*}%b
z1x3RF=uuBJcjfE^+5vR=1`C_ny^>QnOOcvt`L?=HTo#K$xfJa}V7a&_p^FpVSHfz0
zkA-dfwf6-z-%$=4e1v{J_T8VoQ}3+)<$x2v=7K)4FS41Ru!JU6>#NGVz_DxB-XxzC-9n`n(sp?-NpH{i8q
z_18?^<#-&S$XalLMzO1-m>Xy;s2QD@rMV+k*Gc_YzT%&bq^HqBj}H<`cnh)?Pb!bl
z^?9mmz)Kzm1Y(Go|Jyjv`OefuS#sZ7Z7sPnzwGQqb%=3k{>SgRkt{ZHOJLssc4&as
zY2`R!0oMk^kPF}2)9^$syMnVDz=-*Jp*v_*|gbdzi%jMA^EL;wk82UBF0MOHXzmayuee
zLv&^G)+|AJU&upRg3wZc4Z=k%y_#R;Q{aM3#u8z`WraRwxA;y^oFk!zJ6im%t03
zgVomku8-ob33_|HT&)|-dSd2%Kh$TM$<1r^e@{RMcf=H?#gIj0d$c|B|K*;-|3#Ox
z+n^u;X#|dtlF!4%rEt(V3j6iT<;Ga+7vnGBY1d)#Rdi_1cG_6fM3Ik_C2=QpWJV0c#d&S}eFR}2A7ypDR$-uq4n
zaiCQ9cUA3ZFLsy0bf=$A)>hU90Vm*Yw~|I;2BZWmuKd|Sy>H#)i1#^@c`?Hic!0tP
zYxJ?5yYz5A+;waU{e(>cKaOWOjbfFyKUGCWWUu1NP_^KQ0M&J9BrGD?t~{E|DB^c@
z7@`EZa9&Wy2=R=7WLgvyMHeamP;Nd}!~1Y_nQX^H-R{&cq@34~1vMtTv%{<_c_aNk
zJ)8#;l%T&ixI-r&5JA$ks)J7Y;<+*p7ywzpb6X7Q-n756CyaD|-Oa@S<|03-ROt#t
zbh`~8oPC(C4nGRHN!@lmRKkeO)kJ{ofB+B`Amw5|%9TyBEH`RslmP;TXcrP9H9aKT
z77NiY5JWyan=ySHAhkUOQzAz}mbSj$`Dmed7@3=W+_HO2@TkGXGjrDej`e#i{hciW
zBEm)?wfdWJTeZ{%i;Xmoco$f;CBQTyGO|bd)k5G2SY>vDG>i0v=*@F*wnu#sy0JmE
zJ_BlYa@7D8rntMP*Fw_|AN*=xiM!fbNcmmd@!jwtXp|oee3pd@lt|
zNy6|=CMy(xc7qhXzW7^0w!tUy_!pz2xnQn1qldGNa4qlS)qga}ffA#}a0A3TsPT}c
zXt!8>se9?D?Bte6LuCf+@&vYFpRh_>6tmD0NE!$U2_U*J&P7OgKgLiY3D8Y3`9gsZ
zZ>#g+1}H0dDjdYeMUT<$S|&ifGMJvJrX8nlr%6uL=?ivGh;IIgntB^COW3YsT8U?8
zM5lV%Ce54<6n>UcFxSdjR*3>F=`La&hJu7mdy)0tE@t{$K~qHe8fn3K;&z6z-P@-pW=tTvGQw1axxv=L;K5n2#{B;RH-K%q}Ov@@OZ7$0na6?JOR+YKrA
zn|OazRRxZGnbrE&b7##cLdxyqz;T
z9lIE`ox47(#$$K%kjSKkLQ5A~YRYrdOPXv@tuaNxQuKX6x$f?jCGEkq$M4|vu}*x5YGkFU(Ko&mH6Ho)5qqHRwB*$CaN#DDM*tH0e%u_N~lQ7+S(?qNlUT|fwTm=
zf@^O#8im+d3kJy*ba|2^L^vXHsg|w-awIumamP!OycC$h9(D$v1ZcOb6RlWY&ugv7
z6otX(tXqU;B(IKCG?ECVA#are1=#eHP5@2AUfPalQ=odxR5=My3VyNgttvt+v`?&O
z_Aho*5D#Z*jxPP9vYvE`D+I`CSa5yZGwkh%(R}=wH!y&d2?AMa_;WoiWW5-4*EVAS
z+hXf3o&hVso1W=ooOmj(SRthLEap8pu;Ys<13f3#i*)gPsn%nLJJ(&#elGmEsh*cF
zU^sI)Ef5UXqE{8k$XwbZ0<@%t7NwcNZd4x9>EbZ3)9!6f>MbS~l|;!qFGVaQ7gxzD
zQlA>7$bk-jFd#VF1I^ZDmO7bvsB|mU9DIzyL+DHp--FmU
zH}=R$*o-L8PFp7LhCf@s3%IMAS4{o*3$=#echoy+Fj(w0*D(4!i0|>=WkgsMx)`&^
z6%a;1loid`puDZ7v5zRYqmRH5fbQIMkpofwgQZ#YYNcYb2=#x>ZlrE=`8-Ips#HNrIQMpw*C)7V*eXYXwfX@1L$Weu6`f$wOGml_^kq?F#WJ|-NjP~_XRfEZIM_2eGli}L
z@ljz&`ICGb@94)oI&iM)hr>3fi$1iN&sw~5`v)0FmyBw%)DL;<&0V^CC{4iNG%76XUx3#4v*>0@tsI;Uy
zz_np1o-uytstR?URYe@<{${RyxA8l9qqZ1Xlj?8G+;xYFx#J3z@vr|lZBkI^c;~OI
zB|uS=)H*FenQ82OGS=eKKLLFf+T*xjTcdTlZei2-N{qOPv>Aik;@4FxXD+81ac=&<
z`?L~X`ylE02sOx*zZQ6U-7gLH%Dgs@7E4|ml3vT_k0oNmrL
zYF?j|KELomQJ7P#@muCg#OJ3#)N~*u(32;mk-^`2;sk=Ccz9czI{rim*?2rffr?wP
zs)E0qA;t5e=oNcX-=LP;Q2mU4PC_7ggcO7<(oe>uk0=Lj_N;vfexe7%0nL(N_P6-8
zr-A+ao7*_m@XrCL522lm33La`j(_V=VjAxahB9Rc5n>
z#0^{wcZ!{`RF)W+GO?Yop~j4(l2?DC;Rooi6_g1X*>(RucQ%A~0h+_;Y2x8MwtIO7
zTY2i?kdj(%FI!o!oNH)Toe0yyPFIhSl@(|J4)PEeB+4-Fp-g-tH&zK3?(B#@uyjZ$sexPlVN`ZaGjPV*1Y`z5Ac?-gAw
zlt1bdhsSUv9DZrfh@|mbtk=#Rfz8>l>}hIl^i71Jbi%qlaBye11t;2$;k!7^-3pHJY-j1iPM<2Kde0ln&9}w)}!{NLTM9a|6pWUO1A$7f`;bl)e6;J
zP~F{4jDD)fXK&n@8ZF=vDYCX~Cy>bs2&G_Ao?p^MYLdB(=RKmYKNaZfhIdXhh2s-!
zg&Cxj+}*LqB1ALDqA*q1qe5NlB%|k55i4#A1($gHJ}SeNR-S7sW)D0h*ts8sdKk
zwQ4vL;3PUU9IA*!wgNF4wMOmS;UImV%9zt%8KG|oojfe8c;xLPIwo|?lr_%@W*pm-
zFGX*CV)Mlmx3Or9c)s2%DN61tt&Jru71$1{`aQK6^oNwi`;U3WtKoG9Cta_Pie(Z?
z<%_1hcniXm3Q?04HrMk)=mhfl^t+!nuyu0GbH-)mB4Pxwy72n`muaAOnS`>k?bT69
zSfOwhil?@!wN1te!U0dRJ_V&UQe8Nx-}{!B9~bw!5%<~qLy5-zTe^7qn1-sjbLx7S
zyVu(0xN9m0ssgDfDstBoP%8{|NL+kV}H3mzc1n?@rCxr4Z6ARXG^db+OGdq)l<%nZz
z>-E?a)yr7A<>yk_2=bXl^tNiPCtPf=ue?)nnqslOvE^;&l}Mr8i$)!5IK1oC2Yj;A
z5@@T28ds}5c}$t*z_G-|50W^EDD7vhbnv5N!R+-zb8x}l{a>0sx@gBn
z#XGtzSU~UzIxI<_u9MNYoSm#1ZfYSq{5_I6UeU>*KKm#v)Rr0|gC&#hXPhK4t!I&2
z9n#C&r72HmMTge!AGXOaMiot0gyOQ^svA_txv8v{rtMcpMRAJMnxQN@;fn@#PIVoM
z!xo3-0s9BlpXZ17bKPFZI6F=pn|vfWA#-xoMAo|)?N0THc;Mn5%ge0>T3H%`c5<=)
z@YoNab8%ds_g~-W6-r9P`)VFB8oXdnEbmn7u)JC{BITLKlU
zq|I1K*%1-?^cUmi@qN%#z;_Jl`H}$J(?`I4xUY9+VvWkQu%~^QqxSP_HYRXodsZ#st(NPAT^BoV{a>a=`D-d
zVqvia)P^0^(wb=kFTy1k%0}ss0}IA;D2FD>2v!E2m1qJs&G7;42Yk?3D3%RzZY
zf=+NS;!8I}U~1iSZmN2{%lt=u;b0&QPieX3)+mmjU$CXggyw_AKV37{=jpjFUoUd@
z{!pf$^v}8~Hh#U+RJrPnds7&89?~Sm_o}JIUFv?^klRQ>Ih8jq-203Oe}uoa>X;cf&<42W()>dA<$L~-Nlwh>w^dk*Uh0Et9oECaK
zhVtbu*HaeTo{JqF9}zN*FM)Yt?e2qF%LI{&ctv+5#Kt|P2#*DOrRpQuK6SWIidd47
zkffxKk$a`fQ19gz6{cJmmS(EB9?h|oaS(};?_w-FLb!(QmwDVOkE
z{Z%v4IgLFS&8{acDM7z$+
zP0yvWp41OlaDzw#${54K8-H}`Ec6mo$EcNj6W+5;{;NH9|G~`nHGkZ7
zvc1x@kZsR4^H~^V^~xW|h`Nua0S+~wkX@B9z0fCbcQjdp`kHgr(2{44zIe2Qko)v0
zM@pb+@W@2BM>E`X+^bK?w+E3f5FO3QAo4I`338x
zt-;S(&RcU7#ThrMXEhu!3q!@O%T0nxc=E{io7NXPWt?7zv`5YKX^#@sQ?-fJ)a)Q9
ztv6htpeh_}slD7Cj%(mZFf!6yN|_kh%KvZPA~6po8^av~TUo2G@7qACJ({uaZt0L}
zjJo|lZ$R4$kPG3W7rs+4Hfy$kXG}4(6&z}O#yUmZ-%xpa-%aaTx}EvfZ9L~N`!$Mn
z=bQ4@4fbUlFFWG4gebr6S7D!xs&=P!f)&k4I6%G-qOI`;im|ckYN?r})eO
zgLk0VYFZR2=6SZZqDE69!@K7=5S~?Qr8Cve;KCPEDI%MfB|sUsZC+Tz=sw$j*$uQ+
zpW*_r$(uJH$UuYkJd8MfF058qc=nlL8=51mS$MIu>D$R#P2A2lJDW^9Svo_|c!_&X
zO$yQ$#H8#T$#Kg7>S|>i_3FQYahYCb1Sy0hiebVc7zr-2DO8}Z?<{w5aiE*1KCbAr
z0(un?J2gfD?>w;YJ!ELu>#%v|<^5ySN#-F-qNVk)
z?GD;7=~mN9+kWg#sLHF4xj5k@eI!ziQ99IPz$a;l}?fly#ZRRDw-2{>JH_i
z>Rn};6TNm(?Z;DU4ToX1<}u#cIG^HzF+5py)sqpP@3V&H7W>KaQFXjC|H{zHR3LFS
zg)XH7`nazh{x&B;D)aCU`^CB|IR}tt7IPR!cTTX3J5JRb{jQobSg&p!_p4hCF~q0B
z%A}XrMmvSQ+@%zz^Qi7)vVzLeT>sVWulS&fC)h+D9mk`&UUb^1fwd!B`U=l~{hM;R
z^IJ~07v;wg`&E}imLbxz0IK0J>S6j6WXXxm)57e2pQY=1Gb8EfZ8z;Ko_u%M|Ly`f
zn#!$sxXqzA%3rm^4Pwspsc5W24u0=6e67SZ^!vA)oUmT87QfGK#BBBXZ;e^OjcTZN
z0DBs;dW3DZQqS>t0y;zx#^fNpJx=7T$1L0VK(MR20)39$+CVtK
zQ*-zqIm49nJhu=G6OMkV?=X~@dHO9R#@+IHggtyV_u7KJS
z5EMLoht>!Bi@#>Ls|J7)VLdwHk&gY-pTh%BG?h|p#7D|JIYYjdHlqD!;Mo<&H}9tS
z`);6RwGZAet+`l@c!@H<^8vgMei+4MXW_j>5^tl@IsOtcPhC{)d-Ds}6Q?TUm5wNI
zxU#w4e4>1SEsnf5t+}VV
zmX4;z_eQPUm4#R0@DkS3-#aa~u~Xw(ngNT8dagCF)F&?IlQRReO`siS11uD9tnIWz
z!-6{oH+$|a+FP_o@svskx#=2-9$qL~PVPbsewsg6^sl6H_i6-ymlulh0_Wha?#5Xu
z@De5panXpNfk-ov7)JtLwVDZ`LGmt}`YjWNwliuXkT&+PyovXH#-g>Icf}+%Oowoz
z+#%-(szYFsWAd@SBh?hYUvd|Q*h{0HWXLF$|EPq)PDm!^jaP43t2
zX2!YBY-_tz65BkbS_4YoUps7WLfG@dRYaaWhWV@bwvY)z>ruCZ9g}_5^66x$p$>fS
zWv$DB5IyJsSLntY|75|DXk4aq-u2Bsf=QTeUvCNWmQ&lfR*<}_%+CRxl9WX8Y#S(K
zs|46S-C}199$z)&T_O6N<JHtwFXXrw7vJyrNy*b#
z-fv3h%le<4nuk#E=JfgTwt%ZFARqp{wsXSRkJefam;H@bV|7P<`!g1aEG6Xqz0B!KsOP!lM
zIYzxwIH)1{&?9NreeA+89m%UWf@481_QzVhuMn~uA86$gqFCpW1mN_9bUA6m<>pR;
zuEFWa{L56qv!oA^7RkX6l2{VK&ek72%IvBa-Gl~b&N{}Hj;LuG+;3Pe(l#Kf4$AT66e2qE2rPw1RHzC{T!?>dL`N3m$D^TtMzJnLR;tf
zg27+nRb1C+rY`Xma}q(95@omPld)O?z?Llr72B<0JDIcoi(N@kA2(~I9Ee$V9pbqo
zY+F1D
z8)WHI8_#ItGALT!KpHf0m@r9?p+envnA~i7*vmUAcGy#f*WmLZTZ1jgNbDdvaob5v
z1g)<%Gqe$%&-Roh{nh65v8XCHsOk9ajfwV&a^f1$wvx4Mgp~`Ds#?8=)n^>4(gz%ii!RhSz*0hBM^FLse@xtEnPtW4|26Bn+H%Vso
zWs_@WUb&o}{l|^(VpzC=D+qtk9iZaZ*r9$7^LH9Kv+Lp|GT;YcwAJux_YzoQVAtoe
zmuCG;hn80VGM$$2HzLRr-*3eR6er&TY13k-s@{Y4TJ4t^i9LeC!4_e0c^fH>MMHv`
zg*XeO&8AJ?;T(09TFh;g1%-vAG0c-sL36Etx1dG%tb?qkULl78?x^Bb1N$i!9@umqyb5rh((z7^ZrKL1KS+Igsc*#+%_H&AGUv-
z>8InXh=owEs|%*_7hJD;ZLUY{p7*Uz4_$`ggA(yE3LF@i)oPf6+y`_C^6T?gJ%!pX
ze6JoZ#h22Lo(;A{S9-Mrr-v#9Sy;i<>yIHD{%?!>y%wLojY5hfigJOg3&g%864rFgGz#fGv<$@@if_kTK2gLxZupb1K>vI{`yU?
zyPx8ukA(nqfOgC|0P}mCNXJE{FS>hjkhU~y4VTWq`M}-j1FPhGxjmrYcJ??i;A89K
z9o>owzwBEp!>tv$ABvueha!wiOIX*KY&b(!7F}t({kukPN<#tmeu8HF>%FDlIYj;1
zb;LOH@iInWX*q2O@|FKUE{(I7Sh=&uV40pkX{ZV7tvNwVXWSHPbd$I+Hhvekvzd=aH&5XwV>L1E7!S&k3EB?W5qTfwK
zw#d?54kj$#tiIk&E`udMopkQ5*xjmiJ`~~tS>Ee4Lu;Uxd?m1ruXYGuk
zKs!&$7%#{eZW$I4F^=qAgsB3}^JlP6;l%em{}Ki(ei&vLdtjX6={Y)kxqQkoGqV*NvtjzNtkpImxHrWbe|xPwiN+K~KQ!~B
z^HK(J5JvO5%Ezbi!<{Bl?5QPReUaBf|9X3>;@*W8T-Dy1?w_P61V>GEeZyl2nM}-@
zbsl9~t;g1H0;g)njoKLzPOd@pFsPUpLz=n{=j)Q*Ca%tH{HO8I)}RM1Y`zg_!;$@R
zXs9qjz8TK1eZ;Rx=H*EAy|no`VsYoI&3LVUEWi^=7;!U}-*)Rv(F8(U-JyhvmdazL
zEW&W2h)GypzbY=XxHg}-VX%qCyLN4d_Upr@uQ!c4|6`08Dely|+9F6YcY${(#^g;@
zizGWQ<)bafDe|HN8mRo_#pGm{yQqsUBhJxi7i~~g3FgMar`{Vz8;$ZzgVAi3XV(^R
z!`JP}{g4Xq>9=mGnZ`kZ!`67e5&S;0m$(E+LhPU^TMS_o$5V;gT~y0hu}Z&+X7+`u
zuC;PZ#L|&1bSV!GfvNj#!fk?Q_PJ}Df6VILnv=-HUhb~K@u7<4R*GA2eGwEs?B{9H
z-3HhZGUNq*X@_+yOJT`k`|RR|z$INkoLED!d%$Bnw5j#D;Ix5oH>rSRd=2U(8f36Y
zKe(;``0Ql{JoN$vTiU@Td$^N3G*lDGGH8@Wyqe66jp$}j?b
zy3)raZ(5#74*tv8Wb{~Z00KenS&E@r%sbZBJjgwsoH0nSt&o9zvXmFCLMQ~I}
zcfu;Ur_et^AZ*f`vL^{Y6+Ni*7pGH@2lR>J*p%Rl95&we3#7L}9P5)}*j
z?QMl-Jg}V4-r+7oCYH>4$%SNY#NkO352^z
z!U#~>XKRft4AO?dThb!!;VL*?_Nkwh?`w_~UGBqg$2t5L2FEi8F6b6ngxzi=Ey2*k};^I6%@@ZEQ0Vr6HI%s
zm*E3)4?E{Oz(B{V0|N7Xhq5$<}GnjP-S1qrZN&ZJ_JJ
zfvn~u*1mCMn2{$*nq7Ydq0LLqM-}bVrQ$q95p{K@$_iH@?x8J=cS5jDelG}QDl*GoZ=zeUEKS(
zLO)*)4c+j;2_H5?Ur=2Rl9AO;o7W=N{T1@UP%IID62>Oxg>5{v(}(vD)1riqM5`Rs
z4D{bZwDREQ2ya|m(8Ws(2v!@%c{E+u?ND0X$F(x|+LBmRCL7|IrVqbZNf*60=Y3Vr
z!F!lB5YcI_qLQn21wN(+>1OC*y1O%UNA{Johlf`otWnuK#6;sYyL~}w5&q^X;$3#M
zL!O?arQyQ%uK}7#&SgO>Lc0P|McC~ooE8jIuW~yqbzbkSW{@om%5TLIKJAT+hmMnJ
z?0=Dk52h%1@V-hWe>Tm}b=4TklR2rBv54hPK6(rV{!w&Z8T1i6`|WFQl`o}lprD+W
z-96%gj79#oZK^gRB9R$2Kg8|Y4VTGcs2E@eHQ}^ldXh8XZiUajMYs{FPdT`UNE)Hj
zUsi$2PI=Z@`ir$&_;PBgZ?ZyulbXVPs|n;n2hv?Mz+9$3TalFvOG>QuEJi$onTVev
z9Yxq3`G`m;qujI$4`CU0aHF};_d#RCm<{T_VDGWixj$>L9q`!hs17Du-#U7q@|4bh<7?c)USDQq(NIr}phk{I
z-?NS(7ttMKblY=Eb??)xLD(ME2$mQCPfhcw`gYXkgd_bkFp3=
z&iQxBHbo(`-i6AYn(YqA5{g|NujdhW>Z8%vScHyw!A~%yZRrXwN_ON{;>@twnLb{D
z0vG&R|PhXi6L#3*QO4IuI#2zZD@vJ4(ePA{k}CFs)nqFJotBx01MN5`^9a5x?P$y)HVrwdsAn
zmAxByiE>D3w+oUx=IxVKk*@3rO)s!vTXOJ=U|njrSf7ZTRIiQ5q)hqiMjjzgRw8EH}(=*zAeiE`lPRleyz
zm}{JKHQyEuiW+yC4dBogr#o$>Z|ki5V8X}6S(0$uYEs*0gy}!lIGM5@T4a2%f5{nJ
zL$C$NTNVl==B7f^VAP@kdn4C^oSa|TICFnlfJ(Kt!s3wMv)JEHz7SR?WD(!f%6lLz
zKT7#F&47*4)j<|+feRR9U&1ioE2!%;C!Q`(o}ZrMQr1D`xQQd|lRtH=SSnXmGKw`j
zNsj8jdG07bdMf?0ftzYPRchE~T6JyW7HAHZhB-F^#g3he`M?j_ED+XaefYhJvAV=Z
zT5A0G{T8oc!hC;gPU-*a`DRp{7HZ;8bodZ$E#GEPvNdUw@3;kg@=@>ybAtpVYvy@E
zhbC)wm{%YB=R1A?V+N&O;b`0*UMz_o>wL{12mf8`j?!#RKGFx`q^10V;#&r(>NfE)
zuV_G^LXt1ol~6DzzUJH;9HaE0{L%znkYYr}_J4bzHAr3P8z}xo_boTlgU(!?%yMkA
zx8FKJYz@h8OgURQt}CD
zs@a(ld+wynbYR^@6Up5rpt}WektZ#ydE!&2%wO^Qx9xS9*!V(V_}`UC1f~_
zi_7+eFD!2Rbd}yyLZGQD-yJq|XFj<1Cah!r?5(_2bGA@Ln*np2g_X5gf#Wg=RJ`;z
zDr#1q&dyv>@q}OEWO%E4dQxsi_iAvQ|LW_%%!M^--WnB9kugk(|k~s^~&D!?ek{Jyq
zom#)J87av`(S8u#7Nk{CN3juH`*+G2e+2&mO7Er8%{PZ{uuxZngoP6hXgciUqEItk
z%ei}Q?WOb|#kyTXlphYvn|c~p|KiM#icmKYZH`GO$(NWda7dC9&^pe(4h^V6@f)nZ
z1U2RQWnjLz+4;3}LLS^ok3#tztWXm9eqoNK5cU@+(=w5$Bu8l)y&rbcW>jWT($`{;
z5rAtnYoTACz85%?yoP(yW69T&*=Lo%(g1N6)Rge~2j<$B=%=n_byf)oT27&;
zzt7=?ehQ^+?|_s4hDw7@`}FIfS2@p@&a?DL1AUXz|Ago^XT|DxJpKsmx5Wc20IQ&B
zn_8O2-pEHZA#HFdAz_O^xXj1`u|tBW#VYS-8!lDdd{da?X!~~kzO^GKzyQ2|8Q@N_
zy4|(+N1rG8(>sX84BgM
zn5`rdpgh(rbTb?eKBpvQvkMh8(>n6xtLcsnFmt+6sPUA
zEF$E%@!OMi$Fq-D{9z&G1Bl=g1pLHE-Yfa@^)J8%20LQsflHB^oa~0Ok`I8)IY6_b
z=|1+G34Snd!#4{dGDu8Bs@2E=Fd_XrYg~0#r?F1syiDQ8p+pArK-Sb(;b&P1nV7z2
zn<9OH7)GjK|Gur`jnxc#B^ee@GUL7&K<(@p>Y%z+{bk>gcHNv+H@|vSPSQY0$Z`KH
zOGXX&jJW&u-=!wGT&-D1vgZAAkjkYPWm~2HklV>b;#fw%qcDMds9@Jl?SlP{SAXY3
zSAV%Q+k2nnF$yZLkyiY8O
zk-_PmsKX#v`_}t6qx=GJL5YINirMdS*g+D-7^@vbx*N(r6B7Nvzd0hwxGJf#
zhA!@;+=wCA?VXftP`cUlKM!U3M!{DCbYuXs*=hjT$3(th^ih4dixcQeg;B}3SJ
zf9D$+Ic0Fde9q--ZvZF*;AH#_M!kZPP`+{AXBCSn#m$hVs1@7GOQ~|$iC=X4+q5V@
z4cWa4EB>98SnS>q{^jF*BHWp)`H1FJvDaQxkK!D?kX%k_j^dJ=J8e)qBoBYggUQ^y
z6Zu+ZOuRp=V(tWe`q1*rW%l%!u2YGCjRu32(7cCy>)?MzvTdY}`?dlJygj5$lrB>t
z?*K-#Wy0>!rvTG!I9a&QdCsM?X}y|#bw$B9FMbO4wX=G>GPEQZ&(BTZiJPspaYaea
z;gJx_M^<2eb^eYGt29irvBq!0U0}_*a{fVOgn!D*
z1|AH=-RL2P1e#zjJE30eG_Sjo2h-k<8F)^Wi}YoM{819}G!V658eb=#2~myEo37c9
zJ*2q01ysAZuX;BwGRR_LsWYmP$_p*R#IV3i_mLm|CxA!#Rz`8;>uE-NZ8nwv_Wfar
zU*A_LwTl4<%X9XRrEXX!-%L!*eN()h-4qszo{mm-3G~nGP+y5Q!VS^jGPRRgTc5br
zs(}p%?$WgNsZ1&=X
z{HO+pJbXCX=&s1C#{b;~@JY(KD4e9wGN~^tQOCcL0=9I2=PoXGx?Q;y#b8tXHp-ci
zzV$ydHQbbL74Xh?s6krC3Uq$)cqXsFA|I0%zvotO?2)E~{mOok(Ve!9h^AQ>b5#g`
zO<%0O_e`ut%5S2
z5Ugy8S&(@xw_%H?61u?=0gMs7i!A^a9V|ZQae=M^Jm-!<*fKOck%o%-gmr=@9Tq?C#?2C#%w$}-wCf{Gsh5whXlEM;*
z^solo;%QlwTJ7TcglGfstdbzd+)pevPZroY>O|p&LiNC^Bube9<6zUsg`d<`Y$RNM
z^IW2Ou)rn@o{*YCF^wA;3f!Vr6~@LhZ)nCw^r@h8Rj1QTs0!Pv-|}f#Azq);iwa@~
zgZW}Q6-d=H)Gp+(2A9VC-!5RAWp80Kn3+dlXB00IxsjzzY?nkTFJ|UTu9U138FIVN
zLb!k6z-)-P#MN-};lo5eHB4U6ao|4z#_J+Z{fn-fU5F4-T8Tpk{4Yy1y3LQ0zwHob
zXf&%P;kEq+ljBKdg9j)_D=q8!Rn{T_8a@0Mqvncp53UYc<3hj9yWO81i5AuyM~myf
zF5K!s1q>&pv8$)m>4o{e97_As%U00nex1$okqZbdGnT1T!%-;GneX50n7Dn=Q*uq}
zZ$>p`@4f4A0SD&mS$CV{`Z96=;a*#>8AevHPhI~Dm&E~n8o|LXcYU`kwL#ZiTM9=*
zo-AYyY8GTkau>1~OO5v>3)&f*KEUO&H8wte#S&9U21NG-KIS8Fb-JvEs~CfXk91#|
zXK5GXTW3`@|>c|zJsy5O`8A~Vrbz>d+EP$?pbT4F8iqV}mU-lzlvtxZ(dk5K#Z2%tW$n?Z8t>!S-oy&a
z0?eAv^foXI@}T0_M2ziQuJg3}ECT+|{ee|Mjpm>G({9a<0%S>fwwdb{LhMpcvxFO`
zCIBcEmt*yGF_hZp5`mU7Zq2>jmI0*|wn;|SB%yewN6W)|d*yta&nfu$H>*xRuVe-5
zggM+)K9T3xC`Yt3h+G?aa#02H2`NN}d4&p@QA_KYy;hcwub0;mQar
z&cCj%uOE!L4QitB#q=1ut*(nE?9?LeJQ3YIuiXqCm5YH)L*|dB
zu1nHl`74z%>Ma8+#B`0mQ+UQUmY{L^lZ!V)%=#IJq6ZH$nTE^ta43QU{(X7an3yzu
zS|URRu&f0cc6ucL9$Xu7h>*mZBg=okDNg$~t=!{Pn-(8s&PGD%X`+RP8zQzqvWBc3
z=mWw{@?Qa_CF@RJy=yJ*fTkVecfwcVMW-?XLO<`y;kjuv0!&Kg;*ZsZnuR*k+&
z+-}GoJOmb(;D=qD29eJ-KfCe%+nk@M`{`ywT+&voD=*)0Sd61vpTHjT2*J&v*GIdG
z*ECxKJV0)c{Otm}<6M`Q)3ikM%=8G!lO!SUNvbkUyUMNJ(R;G<^0NiZaZxcFI;{W27*Ypf^>;=tE59I-QC?ohf0^Uw8Vfk(nBj!5<_=)
zHw-Yu{l4Mc`&|DyKAvyy*|GLo?|RqTYx~_gp4}t7K>m4N@Spd8y+y5vb7QnZ+rkaF
z;4pqUqEv_6{BJg+!w&>qU8cR9U|aVA`~Dt{I$Pen>gZ@f^)$;_wI6(}9X;}HxAIcp
z5WvSSbxO<1Ek#bZx{i->H<}n7ol;DZkuiblvgJQem_T&!uiLKiG>mAms#iYqw}Ufn
zRKOn%ojKJW0^pvx}4=;Ia#e9{5~ryybvd
zdg;9pAP(iKmwz|LS?6HHw&b9@g`zJkpc-%T*73d@ChiXKJcWQo<9l+8%ew={Yx2%k
zo2=P7tjG$zc?-!@nra3Z<<$)&A!?SB*by48TJzow7Bh9yUrVjtgC~iF_x229FC
zw$aKl`aF3Nl-<-??%j6UB{|q&U(a-iePtgbo%qMW%I9mpTEafK3&F3(Xfji_wr+oW
z<<=SbVJKgleiY?v#>iyuq5R-}`_dmIf3uO#>W;a~ALO4uHLxDC(dd)n8!aPmC8zGV
zHQr=N*b^XGzV_9vSP~q4DfMZ?X~ULWl9S&k3E-DndO(sK=Movatnq=%B8WdEBKLel
zIaI=EMXk7|=U)QgHrbQ>KTxI08Q=R;J2F_E#nqGy1{{pM)9vW@uYmia<#WM*Ti;KM
z^5#EM*{LV}pl&cj>1T0z_XI2jk2R>i=$?hNaFvftg_2n
z<5{b_e5THC$cO4&{@b+YgOcKYikIxO_Q$$BE@Ld*&+5}0Swx#&m>PZ9dHtHH0x8{w
zS8IgeuGfzW#Wp}9y!gJOc|LocpVq6NrIs^~hgwCA0+ZI9Q6S?XWH6bLoI?2|fC$UT
zJO7wj@v5kx9qTiz?M4EoK+4iVs4`IP%AWOBIJAkF5(R70-cdpT6E#-7R-WG8T7m%5
zFli%}pKzZvyE-x%C2P#EIQV#)UN0o!O%nO$(NQ-4AOmDWU*bQOLH{H+7Eg5iM)g(3
zli)XXBr%4*EC!GT&NN03cG4(mc>edOhrVVUDZcx>mRh{e29mm>0lmZg@K&>KeQq;c
z*@yzhaXIT5`{c>}l17r!m>kV{NjgTP|L`Nd1Y=P`z9Yxl(Q65z}kPaoYRhLB3i#vabNE!Nuz
z`ZVWycrOVMhM^I0{1Yv7F@X}AgYD+XrjyZ1gK*^$HsaO3%g^}?=cOH1cD-hJ>7r)iB)$72
z?B>ac^~it;(~k2fGA&U(H8F879VJkosA~+0dC%-@hwnBy!hnLMoETNjRGcUOPu+|=&7Dmyyj^vs!3#laviRi-|kinfeUAo(*YVld}@b
zC&me~wXj$bgr>Q(^2h;pKi_$7X(cYR5L)TW`#f;v?k&tzyqFlfd6|NkQ|}q$?{i}~
zv{cza;8kU$rYLGw=2*bqa$V?lcXfSWuH@f;QZqAK?nHzek5tkZxQo%ELM;6Hy|}mz
z4vmSr?tjw9%rBB_nrrUk3S2YD?F!K>>1}^b`*Le~&`zG*yCnbo#I2Vispr7{Oq`(-
zu3i?$GJYD3@H&v;4srb3V#sj}7z8hde2OkeBF35+8iU)*0gx#Giv1I>!~NM$FI#F|
z6D9&^T1<(1j=ZL8$MWB1E*m!*LOpu%FG?)ylb#{%fha*A&!Go@GyvaL-Q^Y{%(vp9
zPJ5b;@%S$yvAlNUzQ^^#K6j|A4vD>APfdY1(258PuDo&m!xa`d@WX0q4>jTlIv-T;
z>q^r&+U0xw$u~YEsL5Am)HqTL50B0G^t}%~J>6}}lyFr{t&u9ke5xr#Re6}O%MzNcGN^*g4=PonKp`Q>=M*zrOn;~n`B
zQijnxg2_HtM^Nht#f?Az!%(fRo@s?Ja*s=hi|5aNg2gFX)(YXhnHfvpCV5E>ad{e<
z?qTcfT*fQ`7%|~4iRL{VEUeVVk#+sOc_Rwv-!mSV3Uk}XjRFWePd-nFf~}smwI=8q
zX&33OWOxym-statf0o%&daAAXMjs#7P|DOtFN;pUsE@JDMLryoo?#CeDkf{*p9m5?
zDn#;*jjGn!0}heRddLvGW!>IaYAV0_Yjt(x^}@g`KJLKVwM9K-{R0IFg5l9$_1QaF
z^=UrglL8kREOVAQYatai5TvkXQ>~SK?F}%NMr3EpQsU-Mb-Di}wJIcADi_&o@z~;#
z#Uoq?9XgeP*U}$S2BJR>kJaihZG^YEuu8w;@!bupX`=@D4*e4uq_D_a@wiRz!^N~j
zn|g(nyJ0>vdmgn*Ssj7tj0peri=!vIqtz%x_Z2=0C%sAO0-h(|hIK_Wx8CEKK_pkC
z`k&p=ynq9Dtv}&1#JJ1V#08&=eBieV;8}U24e2ib@{#X#51r&z-Ss^HO}9uH9lv$w
z`7`n8$(qGfb!>33BzCy&*|4G3n;}dj-oPS>xYP}Njk~Xlgb;q(fD|EkyXL}7eQAkE;uX7Jf@kLPchfqI$99nNG6JL@%3M?B9}vns
z4Gw02(@1bBURTdK=o0=?=W<}Gu3erXgeN=@m}oI!C+B|+sWitbx4iXeY6Z@{n=8NS
zE%A5k*sDJ{0m$gjMF!<9g$`jyfiSzNzTImegaOi%hX2DD&5_~aO0{Gr*0|(X|IbY>
ziTuvTVK&Js&yk%{9jDQ%4;eYrZK;(!YzgeWX)h%Ja`0+Fv<8Eod|4
zDq{T4WD?N%Tz+}`sd4M!VpWq2MB=j_CzGYeXg{*8@W#G-6*Be^qs&yt^nus}W{u%>gRV$)5k-F%k0
z#XD-75qUSX9dP}k&MH3#ImUO|O65HJw%u&%IYq2m8s0{*?t2*O^?gNY+SXrIHf9ou
z>C^);`nXQAEW<%_5J5>eS=xSMRG`%rX<-o^@u&`C?wfj>sP3(FTJC
zJUPBpX2bAc=9usi_uUY2mzfK~*=a?P6&@|Gm_x?y%!q9$>75U4WbU3CpBMTRAb&Aa
zX7vwEK3;l6~5d(-3p~p=}*Iz?nC1fb)V`_B0)#1Y_w{Jk`i`mjCdG<0N$Pp!rzfLdp|K?Xt6l4{
zQyss7`j$}51S93+4%xZpU*Dd_cCMR>cTTq0Shbl07~D_kR{5~(y4@t+MX*~6k)Hik
z*Bt^3ne@FN>H&?RS#>6BW@E8el7*I5BB+#jHbMZEz*}BR1IEZtBA8v&Velj_7jTGP
zi5Dsf*#;~sdZdymMg-DmARb~}eW$RkuSsjeLvup<@c(es%L4^HZ_(Jaa%)NagqUDw
znBlu0#01%YQt$L=+~$o;V4*qcyZ
z{CzIvC$%BdM_fjn8e0~d>5Db!v?>_+!bl*x1v`9OjcdO4m&J{-fg5-;A3LpHdH0FE
zdlB+TXXtB><;8%#@o9Ss)Ta4>lePEu7bpJvMaadJf;aS~T4MJrF9qBqJ|tID{>Od3
zF&LH%?z0rmh|#ItY4;m5CNLV%^2TLHi$3t8E@)8AUdQ8SX_^a;&JhY22fzs_eL?v6
zic~oGxO$3ct-&8HBy!Davh<~xs{ivl`nGTMhFE7T6y$~H-z4ZauK4@+8GAzKR{jJY
ze}sCLSC}nN&H=74#=O8Y%()rl?C%S&(T{J8z9&;>n#<(#S!gIF>^RPD5_*?3SC{a+
zpmuIINki-N5C33u;~(^0lSEI+{=IC)cn@qjhI*4}(~ED`zAA&PKF322v8}{y~P-wFhqs@mftL*}p0`X#JAH^UEy@
zWFatEtYo;b-Utt0ttXWPD38s@UQrAKww?b-dN1~>ekx%;0_YYaANb2G_T0sknJ
z!H$@u^^kbxJ7V(v`1|zA9w4N^^6ALl#mj8#!Oo~4nn!MNtof=%-}i*9kEsLuApoH<
z^w0UvPqh%umxBrCNz%`1lTU5=M%U|0N-|!HO*uVrn1k}q>5ZLXazCnD!M2$?MN{8p
zcUTTQx{p83Wv^j*%ep+pUaABMIu=oS?Y(}dIi5jGBCf_!7w`6`jC}bu*^iXEVoEwL
zUy!*S-fPSQ3fMhKS^%WwHJt>UMqqpmAS^4;wh25h1*xkJ^rJy
z*#QGtkiqjaF|aGq7rH_$7zHhrROIFA(x|M|CC7z#j`7hv4#vyZ7@FLKcX^8DdT;up
zDV4RZ6~}nCg*bH<)s5!~*|ZzxwdU}@_P!mv0FIF<%?3yJ^DABnbh(YHSf+VuJ;7N~
zPX1R5x6}~6S&a1aBNE=Y)H1Q$+%aD61Hs*V=PsZDb59lKQ?VHq_X6n4krBayKxjt#
z<@l(B?fC2dB=K>u&-jyLf2hdGX1Df-pzklL;onOg(>|9xY3I)_+rvzC*B;{DX*aLE
z7)nEvP4-m@){Ma`^_EFx~z(-|&+}DRaMXGWmp}`F9Es-3ED`e-*FZiSZlPE*J%v`TES#8yzE&Eql2)eIYC_!|FK9zC%|CV*<3_ZtHW&=ei>(pf(rVMBjjZ0Pk4l
zHP)N7+-l~OOWfgkIO_G*Il*5C8N4d*F>TpVzX#-he+f^;u4W4ld)SvnXuJQyBo
z%ME=U4t7k)|In2a`p7r)m-|r0^%z=6sj%L)zNs8?nRA0V@SQOW;I#c%Wmk=HXx8T1J5J1_Zo;}pY5#Q&b!1CEm_A0~ni|CDeTVR#eomtAN!-9wN3oW$zPvd5rVBvyPldkZ
zp!;J#-|@}t3Q_W1NDbZ|QXruY#`G`UAQ@sJl^iXPva@+2a$N}eElHZd=c?IOIHZFB
z7XGvu9mG~+oVAlF^$m=pwVb#frs(xXQug4*|ML~~c3zYn)qmg<{z>`X`N!pPx%wC!
zp|7AB(6`ta^9mZcOCw(CG_{hbQEfBz!TI0^+T_w^mM4XjUiwOkSTSwjo&Uwp-{&V9
z7stlMzNzXjcE6jYfR9VEX~zW|1T5CA_b4HH
zm(CtP>xCyc-vdryhd{~I4%pwnzp-~w6wpwssuCqUH~ObmT=*P!^z7xHIS9mZJPV*1
zu8W>?z`+GfjUyn7Et-kDTefC-l}Oe@3J(sknQ)HLM#XAoVAJ2))OSxYOFa;X4X>&Z
zZORBXzge1@+1UxF6~vSb4pwYb*loW-YnXj}?%0Aqg(?TXB{|lK>&yd}-yjMXD98Je*y6Y5-d9?5VfSeM`*B21*7`x*~C+BC$_h
zVT$<`QMrowC9}Uj1%UE}k?@W0YAT{6>qEd}Nfc=&`U?Ct+S^h*!H6MP54;M{?=?lE;hS&Kfc9jI&qudqYf#
ztswv}!mt!e{|{H}ymTc)G^@zqX9|aEx*p$PY=i#tdjL+=U-hg%XrG*_M-is@yY)Tk
zYiw%tK8Ow>nno(Bu>x@bbnti|_v1vkP@K0&0*_h7f)P6Dr)pT9c)yP%D(6fWUz=*UAF(N=z
zE#tdaME(~3+rbJGX?yGrx4jCgHDe*HFWBmyyF!o|@(uS2U;=XaMIoIZ`y1t^d;*n_
zi@|x8<2QRIQ9k7Tjm#z~K~KrPlt{OWCLQ3bWK4$(rp-EWzhA~B)!7EeEx?~3-};gu
z)FU+Erp*eBuzA;4kdxCpPs@_07;1R%VW@$g({nmheNb;tAaoqa>UruEw40$l_ZgHP
z)vuXm4}Lbl4pKAk+E7HD8
z4FFAU_Ao%+xHatkirkvGTb9F#PD?1c8{aE*`cHamKrDlg-vmPY0f?DhwD|tYb;wqF
zVHhO#>qGZH(p5>F6zeIU!RN5p@=;~mxqr2jz=!M5Wfs1&>JS2S%;7VpTHSO*`
z=@Uw&2fb+;;V4%e7(2Ja%d2T68c>C>S!roUPA@T+=nAJ?4fg{p$&lYq+w`u;r*R&%^dxL8Ae7>ovk>m<$(q{^j#i=
zjU8FV&t)*tK4n=E1*)4KQc~ae?dZGeFqX_?Y6xqO{MM@5t5bQdbdH|Zg>|x#r(I-F
z7I8e@8E=K#HGq_8As(Ltp@Q=@9Nsj8t{l{4teIkP87B)
ztWW#(oBpApB;)PWwC5{J1O!fLp97C+1^IwkW0drqZEt(}9^Fvde|f$poBve)6BD;p
zvOx69ONCx}GDUJ62_&86`y5+1YBqWKDacIEA1>FOtZ?G5;NcO-lQ7I6%@GfA0aA_$
zHVh~M-@4>Ujm(Yoath_
z4k7)Swx{nZv_1IyV!#x!#2a^N9bcxgHZm=PY2z6f}f=Hoy5Dc0DEnFr~ysL|$#S
zTW;&oPS{#;W+HMaISZ)AED=o{oH&7fhZJIwBcym$)-LiND8xF{$?{_eg+tpzJ%a{ho@5j
zkTCdYtbRWFFoWjiY$SpcFImxj=A$_7LPvo5>u}FEPAW|(TP`5whBX~!CNzEkn=9Q0
zqg=H-d3XLtx-MH@Zfu5l2=h>&2vA?%7n|Cp??#-Yy33tQ#BGua33_H`-yHKEIN1u@
zFQ|-lJiqsJB|TDk&-{k=foi48aiTBIUBn9uU_9zWfD5IRK4(rEk;}1izG9qt*%`jq
zti|%H^wNd8MxT+EL<90h(Y+PxLvt5n()%JO2u`1@mzU4#?6;BAsVLQpA)puSipcAM
z-%fq^a}+^X=?3knP7@DHPvfg{%|fO&;>AH+POu}fGpJrTAM4pqWU1-k$TGUvK&6pY
z0Ar-snBAJ^$sSmzv4-LDG+nTx+-u}M4xsv5G|PF_%()_*s3!Cno)7d%yZL6S+Yzhy
zo)abSQDRA?cp;SE<1Bme8GEKvOZ+`b-gO#Xb|b5PYRO$YkjRibq^j_Z6nBz5&{NaWNU^ENd+fQ_0K_UyW2POyrhFGZMZ1Y3IY$*
zT@eW7h}sI3enXJxSnYln?MJRyGmfEeB4|U-z5c?Zdd_?r4Tj0u)3uwfnSEhGdT(+$
zCg%kxmne;7KYTD|Xd{5vnLZ>Pfptz-(uUxH*$~XXpOdHY)wfl6nn$~TNArq+R)Tp@wY~trvmdu*qe=R?hP?dG
zJz<;hCN*wBImoCdHZBKkutP@~3uBglY4F&+-w$erH{Ic$Q&6uBB$FC=Ia@oxYEA%9
zD|z>Dabb##q$DJn$^};RJ5L
zjA58`U|w%t$-4foyf?7wrRoszXu&a;xDc6I_u^ythpcZ8gBflFeaCC>78U~PSm+^X
z0-SbAV4%9**9{1Z)G&j5BUe0HdYRDB{H-|P=>Gm2Qu~l6?lF+<_nAZI-k*L_FQ}A=
z(jNQ|*ZsAZxX*rtpeWI4Ze@!s8n>~>m3(W{i@j^hT=me9w~lge5*9Rt_9^6;*V``?
zIY{09lg4ykaP>T7k&E
z{CH;YnkM!4v8M*f>A9sxse@_A>glwLc@%hh4xPNe1Ud`Mj0S>vx)@f&fPpfm!}E5E
z65&*ePXn@=KqZ#yf1M5%iBoKwWQCdRerg*PPjmj1(458p5KcZugc{
z9J*)Z@$6Z&Axn|5@&!J&+Dt>(?%p!cH=JCIlL;8+ze;gXaSF<~o~G)Ux|c-zr}r(p
zKLV9_k*y~-MZxuNB+m{i;NhKaq}*H>YY%^*2J9<>yX(85tYcq5oT39`MZ%2RT7%g~
zM=4`|@Sk*<5T?Z<`4H6(Sgz?cObV9`_`HZthJb5Z^n0!hlN$E`A*T
z1Q*P;F#kSK*2DUHG$b;pJ~;S+0N~>c679uH=LOYG31S8Z*HPY*Gn4pDm%f5Qvrt09
z)TAW-J{#KmOSy5bFR&PCqw@|t#vTPfDv7tPgRee`0-ZJt^|pi`+5UZ8EU7U}?3J&m
zNlEwx6aNR^MZ+lU+ipqR$(E8H3kkmC%Y|=)FJhKcUs6g!xt5Ht&G`FFoUdHvo{DdO
zz(_oh$q~^Hj!>@uZ5-vMXJy5j;t-{hPNTx{IVjgX)FEu2
z!*dl8!WrkFSYJ4DgXx?LRYQ8A)A_<@WawONF#WcBTJio
zRM1^7N+lo_bHPVK$8Txb%;?(wHSu>lFiGeAYoW
zFzVFqZ;j6$QcwDH7jLglp~EN%_dW-j*yf{e>$Pn1IvfX|Z77@_(B$rMP%8|32g}
zP3&+wt{IhoG4CO`>w&CaUr2~^XMT(@iLn{#5cxgEy#Qj_mdV=-P9o}cNOmEFpoO{$
zggfDJjDv;NvPby$VJ^ML@_z$f&=N%3XdHK;2HL=R@}Tp6{BX){O0P3M@2Pm&s?29#
zNtD;Ye<^<}=^()ikB;-VUH50NU5zH>alR;?8>+S`bshoQek5raybt`hrTp}CXf8A@
zwZs(?>d@sDp%r?uyC5yAq_`%0sWJ(GYY5pIWRoP;Q&McV`i)UM^yUmP9
zbiPcxyE>qsYBYhUi$p)L(KEL*2_#;%Bv-OWl_GO$wZKnMWO>^OlhZH*9O+6tn
zeH{Sc`OXAX|04&pQ~cV19EBd}P2+&e<6H@dMtZ4L!%vI4I0xD_mkR&5(hXW#xW%Sc
zY(&P@oj{R0fT`eJKn?l46@MByYfK!^W=3!uJq-h>fQvD^o@Yt9#muz)YpM8VxGx`T
z8;K|E(=JuVyyMcnYMX1H29&Kj?kqz>@Xz11$%hD-DMmk|%hu?Q#B5U2*_lIbOs3xi-a5v-p|A&H?0Hz?%GYpmVQsUG
zCf9brT?Y;|iBXYIKFDf}&L+wO3@Z@v1x?V#VQWz%yY~RAeDnTxHkhfCr`xR#+9`MT
z0xa}xhL!-}CQ#3;@W2cGI-j1$c%Ra=6#tsC2o@-hyS$fOsyeBvr$S%P!BDgOgJeHa
zrbT=tcUv5&s33C{8kWpfJNpBek@B_aOdJ}TyJ=llI$v6^C}gykRBmtPq)Kg1IqC0D
zyzSH&G8EU6ImdHd*XLTW(2Q71aAq}9cNpsTSrb*MQv|lA@bwO+nFmSVi2nLY^wTjV
zAhHR204S}x2l0xV;N4%N?S@@J#0`QU;fmC(B?!Z342%1zLs%t#JX%iCrIq?OWQorK
zngCQG5}4t6Z7z$`Puvrh{u|>9R%#B1Fh{@ek}T;)WbTxW8Os#JP)65yf{+&ShBh=u
zI=X;KHM(l%N~D|LOt(LQqRrL@KSQ%AL#@b@iqHo^_WW$Y_j-*!*-uIYH6exVz4{e7
z>v^-Iu;W?PeAJyWeX}7Yjy;2H!MQ0BJ!mn*4iP$>z=`v!KtGmr#^JfVU(AC3(Zd4@t?D?Sf`(tjLcf^^auXFmC2%fvb5TL<
zICppMdI)*5zPIanlh@?E$QdLqf-1fF*@K4!8T$p$iBPO
z#xzsjc4{4`PQ*6`Ur4Y*?y(E9U|^K?B9gVwG-9z65t(|Zd8=ZQ<>kcdL)>VwjB_G(
zQJP
zsm-DM%VY0g>?g|wEFc7^f7!Cby_SbiBYD0Tyol6i?!6~%wCFNo&v4I2@f5_VBn+C`dM%OvPN#{5YwTodjFTWO;j`*)JSt9>;|
z=8uOEI{b$d9{`pYdqM4Z(YavxOowBgJ<0xMa4Gp|wMlCew)|+aCuXBRHeq2^$`-d1
zQrLtZ?5!TI$)C7JR3B(#P-bgJ!X|K>gjZ_dxom=d;^5nPo5Y9(JFUaV@jzB_g~we2-VCuG(Yu
zu9v4o_BC+fdwTKiS?J0}o#{tTcWr!18u_C0*0Ns>)tTCq9A_7I6;$v}JN(@C1Hg>0
zD7o_4@wHkipW99ISp3Cjd(q-eJ4#&wR7KI}EboZwf;Kr=n)cyq&^~PKAG3K0vE8mgyc$_~Azzima7dxy
zaTvCzxs(d=1~$0=f$|RcxAiSJ~-C(l+ePd&CpCbvXF9tC*{j
zJdRpM#^Qj&YZyiJU{!y8r(b%$_SctZ>*u1
zu&rygmWu3VTuM~e)flkeeIJ4NP+15IP%l5nGu
zm>!|Lr6rZUQ9R|dyOhBI)C|H<_hkUi|G0RhHlJwx+7pKurZl%2&vEu}ems?uKEZ&)
zf2a+E=H#{C6hXE=2t*>FyfBJnz_VaY80CF^9XVhDBMn(z3=&Z8s|qy~$sg6;34OQs
z`2$;J^%>dR#VoX;VjGCy=r@8-;OQ%sbwp8!@HY2eWOAGcQCKMb8fD}j^gx)~i-Z?A
zI2t0fTy@+>(x8pS=tdM^Z8v=-w?i^Lg*INDCSm5ucj~O6@0n?1EV+}`jIsg+p%fld
zf1hm*`t-Cokne-n%%}yK)(^
zjz>?%6FnK@oAcu*2n4&e@5R1qG{wexlB83$W(w(PcGmv8@u#09Kcz-3ec5v(0!m&0
ziQnj-DZ5?{p+5;RgE~*`r3)W*jwi?#&a^H}Kwv}Wwp0<-dD-&MdfEs?e~nB#mZ9hK
z1uT=JZ*#!EG%XvVSxP5v$cT
z#o`DkjZvL{SG9+
zz!_|qK%ZytqqOQ2QDn!Q+nKONwTg452}fvrVhvUuQI@d`PdUVD?@P|$MVdkaD38`&
z;8==`Y+YCXR7g?rxi9v`o9uXRr;gMa+APm>X#!D^xkp*76Xvw?r*lY5^b)1q0jsHO
z#iyk4pPVhkvZp8EENQ~BV*XwGm~j_K_L=IM{R@|Urnwe(y63Ui$HxR7{oCR<>WN-?
z-iw_EzP76imO2_JH>-vOBJXTyR3PNlSe?m{(GU<|kC
zDf@K>>>MuCO@H`lPF(&Kr-1Wr*Q&ZykewYn;{jwiQ{l2YeO3i6j_5KtWwvMXo*Qkz
zqp~#uPNoafsg|W|k`5zv^7_@L8i?<03>FjHzuJMPu5w4bd1uUbJ)|b`|A*1C`UkuS
zJo;?T0%fZvuAH5jT%2+FueUGN{gR`?*=K9;K5NrDjxJVKLREmO|4*XWf05Q)(PsV?
zE{PdGtFHG$(||%eK#&5hf{N`cIH|KppW+_Ol^%itB@H5Ec&PJRtgqV|_I)R~#IjqBlR7jbJnfi(fr
zG2Q^m)eoi#}S?%G-8U2+rR4ci%h)CP4&N{wtG&pF5R1zua*?eDVZuMqcnC``-`NHodEv
zNALk}JY{Q#lOI!9txqHsG(Bwu#%~I#d?yG$04vcQ0&0!Je90vfv8TK1O4QWxn=vO^
zL_{cvMKtV1e>=;EO73n=cguLDxV~_>?Req~;wsH0s>_wcQz_tF#et{>Xm7f;Se!rv
z&sOMKR>h5e*L-n?j1dx^AW!KToN0YmaBPWR{MCR)ffF|Lnk5F}y*6xbIhL+v2iEk%
zh?95U>+S=lBJShd2-`y%Z+uoRcth@-e%6$n#-FAvhP&duu2T?k&_ouP_M+C2@9d&a
z7%ynneN07dCp^1+2{>@UuYo%*Et8pf&~y1IDqqQVdK12HQcWDBk*SX=r5iihzkcW?
zXh-=x^Z3+l@Lcf2>F!nhRtYN@F>8t1C@Wa-FJP2s|qNccLYXFsY}a
z6CTZz_QDcwcta8NuMgO#eMitC){i*9_~N*5xh^Gof7CV0jYUQYCe<#xY{BrR?Dv5;
zrfpg}8z(@hwJ4x*vQ_X^?2TMsZ=}y+EQi{%?t^#4Zd3wP8a0af&h*u?JIx**j^&j<
zfTa_?3#uKffGg2!DBRvBZjBw~C7=)lLAG|1rUjWASSz0HH1Evsz`g>c#F^a^c*ou9
zbK{L@CmbH_M6{Qt>WiGp7m||*6!=b(!dK>Gi4>+{cu5TmrSL7Vv~SP14~!g-^WE`I
z?~X7yvMNGPO%Q*nS+iZW-C^eeQE|_o11EF@eRtlfg}0e!Er;TJXock8(7I=m?7nml
zS1wX|*nKL)y@0`%5`u$W@$~Q!IWCDe^><1d{RIGnCi1WBa@^yPwq5VA^L4v~B(O^r
z&`$p_)%zp>N
zzSjzEaCkoKNYkRU%27fTYs;OGpINhL6k`1B`#=P!jP!*w!ExQgxjbLHW45QQfmJ@d
zp)Vre9&VAd(RmN@>X$@`GdmVS-nJgqh(w!{dGzy~^Ny&B=8Hdwlls^vPVWSsM{}r!
zhfcfkJ{H?)P>W$!GOzlkJnOR017IQ$z5dsz{9+4_zQku?IRW;>;8Vl#LZpy4M+ska
zvLlUAV5TkMd0Ipre2aos5-`$k6tzGB5x?$D_*(jeug}ZQifL5Fr59Adn&A9n%E)V5
zxM39ERfw;)DxiV<#?WQz*)rf<^=q9-Jhk9p70Xvl2fHx44r>n6uo?H{d;Zg?m)LT^
zwAlxmS3|A-1P;`hP~cVMVFg)YI!M>3jbocG`ePT!d9z>*p!eDM
zc?BzuYxEw;3$3MvwV6>vfFgszxLp#|xWF5d#vO6n&Fu|)t+ct|*l9p@Ixn|uYIZpk
z6-9~CrR{cvg{v-ly^0B2hLG*uJ)j=scsE4&JkW8E{d*x!98kCwa5q8AFyCt!MZW8M
z^U38j;QdA?A=|_FH+}A`BKRD$1k}QK!`7LpGI23_7u-MvY>cv5P;Kt#mPuik`W$7t
zZG3j*`KT-YnVS9cM=P5&&c*m(LY1n3=N#>w+?YMAnB72OMT~t(heleUct?6lnuX|b
zO+jPUAq!ELLz87_M
zW0c+H@3-P#PshMArM2-V311OTb(|YvntRG5NN3vuQG9)^jtRG9h>ozK@V~kT_Lr>r)entW;1bc
zTGgB5byfbt3#L$0{W>~l>b1pJik2*L2#GH7=o9GnS?BeYwJ)Gb$jHE5D~(M_JHWX)
z31}c@f3pOOHt}an$TDU@)O4G_l#Q$Dj^Fem&L|$nhHo1QLLI@xeQP%mG&WG$y0&Ks
zvsQ3MSM0p5d)7vQ7mDl^Qkzw={x$V9LGC-_$kSG`hO56#-m4|hTOpQ{0!^6Js-O*S
zbY41faY;|4oGHXkK?*NMc*I&_miMs1><(*FLF$pnMEOZZ(@C+1hZ=~lFxW7`jnL(s
z+bNJVayt#6Co@vRMy+LuhXi*O^-@L{-lVUWu`eOTgrmpM}QhYts6SoG(~`W
zv#%E=pC&X%Je3f^=EEHyF-b>GOhb)AHQM4gZmEgp0vwY6tMX?Hlp>!9J$k}f__Zayj|F_2W(
zF_+(AX9pS|1Gou4oa?rMqgCY6G;~Tt=(B%jbh_+x)VMp5L+U=(In-a5lyh+yB=MQd
zcD5?${x?8h$AYzK{nX}H5fq*mT@p|UfFLmycT56f9L^z)X=I^RT+l9-ff&Ney$q>AW
z=K+{L$P6^^@Iw|opU(r+J+VyVJllE`TYRv`(K|GIl>(~8bNyRfVGU+I{<`#q%JkPO
zH|ikD_KQJlrN^eBmj&?zG$N-yIT693*0ZrMBK~lWOr~2aBTLlmcF(A;^FRs^BQ6S@
zR3hNi2wv*&^VX9CY1F$Z>N+;~6NcJa*eAp0>*A`Pf0{mt@A6XIP-B{^YTC@Wf|}$3
z;-ZfCpCeH8^7M?ueZCozoNg0oW@I#!C!g-v$4m=hm-@ts5Hm9}6MXj_ctbKj+-B-}
zSK(3>EpwxfL7DUEq2%{E{f((xq>M>7z7sl=R!#C0aetryyYl&t34dh#*-rMm+@1-`
zVlvw1b4lY;T&Gy)b!J5t1
zRTA)oUQ#zrHIDPK5C%946feA}48ZMjHkK%;W@sQG9b!0Sd0oC0S~izrT?!%Y@|?8+
z9I7rmVBgE=9Kg}w0rVRYb701CPc%(L1)>lSp%OYB<_LJPk8K{jd*()@A^7l*p`{sX
zzWjYiZtlr%L6r&>hz}rVRZ#$^+hV#PxYK;Ipk&s!D{!LixcEKT7;xCqw9}p1+a<;_
zq?I`dR4K3@H91%LVmrg19oNPA=nm1>y3W|FXGX*HqHRE>Z9wzG(5E7$&FAX3QWTXB
zM{hW%E8Wh-EQK
zalps}S?FG+cg+?%OOMxy%)4X}z5L_P(u`H)l1ABcV{k
za9LSM1uY7v49pRjcL5=oH908Osu`jdO*fsN@Z`6=tP-MD++Q9#H@a92RL`ymbXJY1
z2F{Y?VH@9_sNL$3hYZvYy!JXioNMa|`UO+j9UFI_k%tiwk27Z2yV
zv9H~qj*XqHJ^JkkG}a@=o;Q1!^j{8mbzXSi+-O7~=d&6Gx8kjcMCD#9+Bey4v5S>e$8er@w|&%%W;LiIyx&;tsBR#?MY5B{z+}
zs_2Fh;b23L6+xSub(yCG6K%!7tK#@pYl~R89ohGl^3`~D{ESi5(0|*r6}?g+OR`m0@NhO
z7(W9Q@Yc9?9746qy8l%Pt_+fSaA!()RR&iFC1G)#OBHKh53V{lvbe=BnqYM&!Ej
z*B{GhS*s;)^R!b%a{@s<^NMfEcEZk`(`^m_>h(j%I?B>77b-eTPPb=QV5`#;D#dH0
z#T3~K2|_Mejz;0o$qUzK)!F7Ug&zRs2gCjxw6LQQqV9`bJDlvVm}Tk-``R=L041EB
zuZ(9#vqx;M<+xq`><5el9K0NrceFPKK7IxkgGzxGeEK9%nWO
zS0)Npi399CPMf&1(5iU8WFJIh!NjeJGzDLGR>w@g2}8)`x+{(!0lH7i1+L1Zq!|uS
zxho8o3@bL2M56(}t35??@}Th)0Myct1)2RKi>4uwo15!^uEW;|gff%Xp%DJum_JlP
zzw)WmeL3Z>4{eorm%BYg<#@g1Jbl>8UR
zTreV64LS06bwPBWt%j6|gArJjH`s{weS&B1Zi6oN#Z12Pm0B!qkTv1=6ke&|43psC
zVLEKeND`PQMd*4g5a%2KNnE7a=b`Tc2oxHVmSZ_SFQudF%}=#3H$Eb#rhp0YMpvy|
zErr!-8BAJ5Dgl>7h1+8sbU=G8;)Bg>=ziR^-LPL}Xf*sG8HW`Ie}ERzDN09=E(G^{
zR9Ta_B3>O5oihgg)l<-b`StLWQ9pWYc-ZZt7$vcsMLR=J4H3T!g
zdvB#46oB=c?54doNEF7kwbxA_Y;-mRRR~1|k-tg(>oR-vVY6tH;-%=ubv$!{NU0|h
zsFBuEw(|!d#Aj6yiqKI597mZ7W4E`ZlYy>48`5`9lWzL-<#kz!kG}BaSi*i~6{C)J
zU3Z(${$r%p!cHhUaQ7E56qAH$E6mhq3UM{v?diV-@Md>eml1yteFwDuHN|35w}Q)i
zum|~*Uz(VLH(MtOfUniM_9>h^iWXBQmw8^^gza;_$>&}uV5O!8Gk#-AV+zVovf*wy
z9}%b51gLWGm~Q`uQTP#BE0U|BvhuRS$VehBq+)9tOin|B@
zPQs^RHCJi5oJ@G80sGrt5ZmxYg#q^}d}W5o_^z*O$coeS*_K0l*j_wmg=ENLt>&v+
zerPmK2&iQ{f{0VS)e<@l2c-nsqyVUk2T`Tb@4o+A|v{+c-sW;al)%)#%QJd
z(Pel5l?cTb?0xF|9+>MWc>JxwUL(FTz1~aSd^gwT&RhC!*b9hluDE)v#C=LA=(I=stRcbgDK?
zhj4Xr5*85-gUKt%;7O78YZsf|pY3+=Y&ZT3c(U`w0eRx0WS3T3m14+r?wLBZ)X=YN
zsv?ZT9S!5(zf?uJ?Ipn{F@$96fR&d13u-bwVIBcW0>;PKXDE!dHaBcd=Vms3%=Xs0
zR(`l~p4_T6;$gb0Q2x$&$Nu(~F0}->&?K<$<^g}*bV4xxw|$p5vfHFJM@Iuf0%_`n
zso-~ZHh2$3GrP-2I{MfG)>u_z@%@pSKTEAGm{*dQv
zvo&Od&?OFaaTOSP
z7_6B5Xh|b648PNlZXq=cx>eirqw6GQh?Zspal3~E@dGq&usLrl(tcYTp^=yr0RcLk
z4PCvDt}t3~{Z^`JEN#pc(sj&)BNfeG=GBMrv_lp%_JL@fFAvz9caaN-bNA-2AOO&4
zuexUA;zM?I!T6Szmo0zv$HI4oR%5R+>IVfx#u<*jJY^kEY|`F)va;Egq6}nggR}yB
z)Ern&mv|no)?MUEY)uq0c6scke4cV(8Qs39Wlza{gUfAA)+IHyIYBWO8>&WABj_?O
zCtz*A|7M>~KnU0X3&wvS+ow0BXLmokhxDIdGwoX@#Lth0Z)Sys5Ec?NFASr2
z3yXB6BGnSRCSQw(=4MVvJ5xw^G*GZo1V`39RJpoCABPEdr(CwbN;j
z0iXTO_~FJ7%>+DCF>k^U%Q(Yw+4Yrsu%d|-_X$K7m&TscsIAN?9z(Iz>*~ar$EGQ?
z0ta|S5+3+D*5-CX?)nH!+$?N3MvsTu8
zYgbcTkw{x50IakHUf??lx@fnr01s=eY`^bUONNuaF=1EK)*n6uM@}4--y}L~{#ya@
zcBUsZqlyC3qgf7ijoCgIMk=6_^9jg*U+f0x*TBzHcy+@mr53Z0m^}~FUZ^}ZvfyX`
z!d{>6+#U3~SB$xPVsbo;#Kyg_Ztc^(TTn^^hL_I&y-K)4S(hMsMteL+#8?@W2?(^O
zUVe>b{$M2`ol?wYHM9kBr1_cUaQwyUB;ECG54h+!a3}y-_?!z5pIk-6{e4}>T|n1`
zKB3BgB)FzZh1rI!3Lh
z4sCztP8IR>9vaQ~>V&maKabC@IzlLyc@lW@{l
z39{Ijm}AY;!9W-af=JEwd#i+3_p}&Ul*h|o5`u&HLj&}8prKK)PP#pi@OCyMjY$FH
zqgLP-10qMy&sdY)y*mjfwkqY!I|PhEsvre6&l#Hw2f2U@Fj<9hx+hmLiC!D?H482h
z_l{6$4j`ObO^sXZ?nVgo%w5#bBaZnsLZW(Rat0+7lUEZaXjp>MC!?m#Gw7VEX8!WQ
zyphf2VA0i2Zvb)VJJ{(uvlsDW_}yt!854?Tllb^Uv~{m|#)%;oSJPlLAiux
zAG5E$sTLTRKZ||?)Z|A)q5}%*(@KTJtNsKalnKW3QayG&7Gw%S$7UFG59OiBd1?`7
z528Bl9XXGNn4yOky8^aX+2M8`05lD>76w$tr*9SdwReqq&5pf5-dPEy=47=zwbVzv
zq1U3gP8oEL#R+DRzQB`}tP&K(6N2i!^Tm>DVh?gqA18vJnj`_xINg?my}4uveo_tC
z>#x5|U>q`6B0nVz*qJJ8GTYWgenE-$-D#|OX%*kKIG(}wudu*e7BoFZaQ^mkiC1af
z%VqEl0_#*aE640cNew)7qXNwZB67^4mtzsQnn6)(Y+<7j4}TAg7hmXlAnXsSYq%{<
zT7X5MO_JZgM9cfy!-h?#XVg)uYWe0fKQXJrT9Y{}X8iHZz(cp>hnn>!yz#^*e2;ig%(o&z^o1l!(BXXG=qU26S_@$aqj2EPe^Id*WKfcrjPXck!-r}aXieVKx+9~79F5WG!UI|j|v#j76ek|Q}Y}j2|
z6RiT|}gjr2Mv}Yib-2K
zQY}LFNv3-QtC%8lxR%?}vawYv3}u5wJ8N6tIp?MDAD?31e}V4+Xdmt+k@|VZF59Jg
z(i+drh*u{xqv^(`c+ilPF}o`WLKoyhiWt?CI@7Pv%Ae8lx?{#bofO}Pl0Si|LB64_
zWPD7B>qMr3GpIbLmRizuR(Zqhmj8%k;ruu6bSKlV*}T&Nq*-6&w^6^Zd@Ljixdiz%
zUBJ{JYdS#Ll{N!`Xu+dW&pSn;Q0S@--dX5VXxNnYvX>R717Jf>1wp2ktBNg^d45lj
z-|#AfBI+|oDUi=H`Su=z(n$5x@g66ZSRfdxYB$$+MQxS!ni2*$msO-YI>xxm1;RKH
z?u=&Oxjir)F#(iSqdXv;-Hu90`Iv;?CaDcpbnvZZFV8jWZD_!qo0Uc~hE$Nll{ArC
zj0L#bXN?tWv&#e)hx33SSeXlaG-;%Hy93<^mJf?TKm#Hswt3wvw)FjU+qYQsO!gD6@g&CR;h3cIuTuxw|
zu=>byX|GWMLbNrf11PLp`sy9(x>va{g+L*tzHI7L>tdo6ePKyWa1!F*rt4K~zB#!k4r<%m7t=%K|8yz{Mh&>iSl*-MX^0m1++_D?(LY;(E6*@8L*Pqx#%+DSO0?kLc4I#Fa-h^G2^-i+
z^sH8c5T@YW1t75q_)v5S|ASukMExN7=AGj_HXTDp)~`hUM;Vp*zj6T#qp|mk&fOyh
zKkr9M%k+2i_m7%)cp%z9^bu{67<>&F@B?IrJQ)GDrk+EFhXHTu2BYF?3=7n6fnO2z
z4v8@$9pA|O$HNleZ3-wxMh_-0f<+}h^Vy^TCJK;Ibm!NiW9d4?lPJ}~GogtOq|uCu
zWB3W`!riEiG-;L{BlgZ?8(uBs6xS+&x*AA9K|A08^yPa#z*WBmfQo+ymz|>NvMc7?
z3ET;Sl7n%v^ZBP@J_!b!RX%}SN==K|*I}J`XkF*Bi9^dN!CjVw0Cpq`50AhKx!Y%_lPyW8h3@zC^2uQH
zL%#qV0OwE(b#gkqFfyZYW
z$eFi)@~NSeUQ^sm39nEIPA(a*p~p58OPa>?THQ{vrN06Xq^~-Ni+;wfna*G(B42jD1!wp{Og-$8@F>$RDDc43HT
zOV3^=U-X7RfQ^PDhlDo$Y1h6x$-Qk+S_~RgNo%&`=ZqXz%P}`$6YfIAS<-=+-A4lg
z2p+}3Fd(aluT0H^N}rjab1lp@)6e
z!{U8cW>#>uRP|%d=WSY8(h8-X?aEe6)b?EVBKZnyOd-#P|+SWxaFzr|^?PzfhQ>3PZ3#WiDbc
z8G#I*`61K>!qRBgDuDbyNCb{7Vn48_4dKbBs{>?!E9z;9z11_`WqZ{`K0KzgwNm9S
zkwoIY>(W$J;0&j;dE$Jd?)_~A^p);i?Zqk}&e=QqD1lGF{d5yIMoVze-(F`}oFGGz
zkt5+1C9A3}oLfZkojQCHI#bmp*M3{gtUc}_{Qh2=hJvAbJ#>%jzss43@{RS3rA0fd
z<$@84zKY_f>X~*-auPF@?RFB0IpX02Zr$I%xVm^Q*K4U3H>>IUL|In%d4u9&aj+B6
zpa+(KE}L$-v8-&2hbqp-OUKTqRO>-qv_;1
zxQTZFUa+Wom~uLxep|r3X|izT;z6BnGpqzxS-0-2lZtpQG8Z1z%cT9P{mnN;Y;pCu
zlgarr@6w$`V05dA7+{8$^j+8r;=OomQ9`g$_qjL3VpS|%Wk-iK2&q}0NvB2(x4~MM
zme7*ezw+yy-Dd(r8)$ZQAg>fgw%BPo&%odf6ZPs4NI$>zMaCcgo(fr#%!$8ti!lph
zc?zCKEgQ8!vIiRXJzttMkG&cmZXpzxPKq6To_*CdpdK0Mt)6_QHO=<3u~boG&hGf+
zhj;!?BQWH>1(;^#os+;}woWkogW$Ex8?W;DU5i6K3&!R@^&iFEJnFR&?8Gp}a<-t?
z#LRV9sk6l`>Gvr??-M}EX&geunzVdc&!($z6^E9c7VBeQBA&{
z>4~wO_8xR@um#fa$H7^9dHKKGwKNGlk3`n*zq=E51!4x}r)_Ny{=s_qTQ2$`uDRk@
zg6XVxq#m+`FtPZ(L#L?U@fbfbU1T5LEC#Z*VIQFfdK3B)j$oRe^t{C}5;v{vt?nB8
zIY^Ti`6vpH{=q|OI;NOBF2=L9z2l#cuo$04J~hTaxOt~)lG^BID}K%Tv*j|)5lx4s
zSS{>&bue%Xb5Kbt_yN9{oFaW-QoT(_*&DIuoFscH#c=`&=0Vb7V{vf_XRulUn(r?_
z_2A;0KJN~vkteN?X7oo|DiRW#7D3gu2*uKKen8alx}>*oJY33&`X#;JBhFBIX`p&6
zr7>?3`UkKLGTN-?@=B5gIht!f)fbcbB?e%hi`q0;Yx(0SbWBzLW&0qE5V25t28`(>mb_pl7d9sPO`)6~=JgK;CnZHy|ssv}2vG!--37?5MAyNsQ7FcqXVD
z-rj>Jb=Ax=vhB{@a+-8``ld+`1`PfbYV=kTttRszg>p2n|1dK+SO&P@Y1v`Vf3OEv
zdlop(%3k*#iiu8*6DZvgVHFR7OeZ(*Bb$Wm!NG7!b2K~c%f$^tt+BPE%Kgx)#OYQ{
zC{DA$vWHuZ*JpF4*NTvg=p2!J$F-O`@m
z-D&gH7U)o;6_d|uKd9{lzm%q4WQ^gg^EbBf6%G0|HyeQjJ+_bxo3QG{h#V_>^Nt8=
zh{4I&p&W2)`JzeKlPzi+wFSyy2h#~NW+^
zyJ@3ReWsGlme#NQ=$9|7DNkz^u_w71H5^$qa)@*anO4?sYq@=6#j2
z^8uivn_Y0gvkJcz34*>g%(yrs#iTKMlLRfiFgLgoM3?7;L*7ui0YC7wqH$orsJK^&
zxz?kxQO;ex8o^|zaoanVdu#u|4GTE-_}}+Q-8@c<@Z#D#-K0zk>a*key{g9~vqqf}
zM>B?5#EFXR@Wy>NMdOR+cvENd&lF)-mQ~RvePIm~1mvb9G5}Q;9UwXUUbbLp&smfhxLF$2M6%(s{$I;;X(i3b0k$8$(})
z=E?eljjPZWkAXv7%F;O@ty35%IhU8l2&Y;QGOT0fdBHVLYFQJap(b0X$Q-AnDHy}O
ztnu8F<3%N1B(GZR#DKBFf)}Ez^sm`eaU=W~gF
zogcLFS0u8NdMU)lL7{V})7jY}`!#PQGJ2kTtv67Sl?kUJx}=TVqDXOGE;WpIRbeFg
zFic9SbmslBXI~+3#2hB-BV*<70`s7{bUeVgzw1GnvkwfIE2zRJ)#xaSiWjOQv&{
z=XuT4fZhLH$;ifo?I4y@086T?EM8nNY?q?88_qK4Hno2}Yg@KV35>KGyH_0NobvL)
z$YvNnqjZ`%H+NMcb-O#y7%fCJHR;)PPF~JvqLwQDSigQ8KVMbTYtYMzJQ|?}qz?`kH
zAAtTSa#l1r-MqW;sT1MVD;Wr8rw)te#(E5aNK276K*UK|H3DHIJmQ@uBd9s`<{aUJ
zk0lBq^jZ66WCW1RL4`53YK)C{Lk+pZFKV>6C^$3eMhCO@A4_X<5B&ZU$NL&vyF;0Q
zEmRF+;Iq_!gZZv!4>UO6m;3>SGY37ZI6TAzYGIcyUG(+^u`}yeLh%YH=olvm9yKe8
zGi~v1hv>`a_|xu_dMsmlc+3Dp(E!FSU5ENjpC1Sh;yp7$Z5uKXA#522Uync`a`bXj
z85G!JT=SAYA~qB0<|R$U$!?|pTQkr?)tBoRhm0X=Bl{8GwL-Smp#A(G&L*}n&X^OO
zDYgggqoO`^(T>%bfhP^H5tIq-zw(AtsXm!<)|+Vh#S*Xhu?o#s%Ej!xxC<>0fhxp4c?7lm6u}^CpuWhRNQK@xM9lW
zJLM)D#Ln1(Z3oFF!98|`&MdQLm&I^oW*~IW=Mo(Pij#ZOqjeCll*F>4*qLsxcgGVF
zlt>%=6P~Pm&)E*E!0C*Vw(`(x$(iER#i6&DWmNv1FJv|+w1dai`xuaw#o5W
zHRP@P^pkHczo;|wEHq#C%YRS*J7Zj-IC_$3B7?+`AM@Z2h`!scqUPPh}q--B%6sxBnH0-3>Xisor;uTYrG9
zo%NQDH$h6KfjN~C5;>T!?-L?^PioyyfMz4v>*n0js01Z{6J!h&2La#sm&C2@-fu`wyQv
z4Nblkk1tW?;!E?WnIF9;DZB1@oQzMzI@HU{CitTyXA&9a@BRq3+$J>%S%JgTu)*PZ
zADOyruWJ&tn0Koky|Iw>CV!9R+n!$Ix09V}sm>p8G2{D7cLZbrr9AJ87o#DGfnmjN
z4U@c+)`2}?#>>BD85jm)@0ME5JJ$?^51=|@Y@Ahf^VcTqf^#@}!sO>$2AN7~
z(p=Ek%sdlKDY=B=5~$WwfuQ9q@WqZUr@!sgD5F&#=`$lXAe9k!z}N-Wp59w979b%@Ze-;Ko!kr
zwebA>u^)XYw&-)-U~Uk87L+5Hv@27it%(@+u%V3jTX|*Rg_!d2dIavNJ_N8u&G5)B
zBd)Ebzy2b6STL^yDYJMK3#(29${8oV_GpGfGd7b=0!P^u3F*1^soGT5pSvRD`oPh1
zk-&NPv|Q4*?HNx}f(Pu&uu~^a7O8P!V+YE~L@SA4{S71VmkR7W6B}d5KM9iTce!--
za+qqmFBsMe4+197#l0PriSc_hVX2kMn7Bt+Pnc7?>rN$38YKBsva6YV)4e5UCY4o@
zYT7zk_L0b9Sja1(N^L}Q=G47Su_z8Gw*&g`cUyP7>P)7VsvOV6I;KK2rpmvtRx=mR
zh8bZ8Lh1XZHxr=|wKlUZeFFQz>_o@u^O@hiH8q1wi^lW1K_2a#*S-p8$`X>np0N2R
zoFP{s(4@wx2_B6v=7I;Up9onoCJZv*g)Cn|$~F_7ysKdwhX@i9;t?9|lhY6Pv>*jc
zf?|5WNZP+CWEu#Tv`&mMl2Es%EibyI6iV9)Tcn$LmRFm^g7mXLhH3|vm3179K
z*2h6}rk^0zp)4h1n|BS0?jcU~U=aJ_W`>fvPLvJeJ
z#?lyo(NED%Cx?V>-L34O{dY#^tcym_7IA>R+$?V}?l&P>2V#^H*b+$3ZvLJz4w{Pg
z`m0sUO_4JW+Ryfy3aTU)l2{;sw5@99YD%x0a*q;!3Rpt8aN%cE|GAI&chlVfwi5@p+0$1pzp`e<7Ne^Tna=m^`7PidB_$
zyWIz$-ww$x?TYZ~DD&Ac{PTJwHZ%@h3NtNXBlQcD&&+=Gm8`3S(DRYkwT16{A%R#Uft&$qm}Q
zY`VitC7tz@C}9Ony4J+^z#B=nidrawkApKpuI-$b(Kl9E=hgZ{#hAJeZ*-d0iN)JE
zo4!X_j5n%?bK?6Od>EUNo>QSwEs?G+iIuc~ix3{dd06Sb!dKwYtfImm(LT
zqF?qC@@3~;xNArz>)PfzXc{yXzNd4Ne92MG|)T=PqbT`++c^oN849_tY
z2)3UiC4K`}ZRFJ>>)D{KSvXz;OyGMT(h4Yn*UAha9?M$0-L`j39y{oRY`9xa{hU72
zHkiyTQ2&vda;ADR{-;e7oKS1IDo!~!z0vZ58}z!EY`RztWp6CP#Mp6_t0^A=O&y^r
z%bBf{>cgk*AiR#I;~(~TQSCLp>4wntC(Fx@H~lh1*>ktv>SLi4AM>4@4Q%MbMsu>o
z($i$=5$d~Y(`=b^q_*SKQ4?Q(Wg+DNsEB8@>hsaikx|RlCz=VQ7070#IeZx^;%}o;rCgu;HTWqhvdJLpG|T6ji1;DZ`>5iTgj~Kb+}|R0<@a%5yxP
z(vDf40pHEPDll|rp=6yK=KZJI-Rs_$o#g)ttQLz*n8uG|xq5i|X^jp}`<`R+`oMq=
z9<5gk>x)r)o$8|4p7)t1R`urYGn_Kb`bvh{E&zkFMXRh2xr=!U$Mh0Q+iv^wL=Ouc
z6-K$ro?c#plqe#RO+lM11gvg{%d7y&Ocr!EY`2EBQ|6)Uaz4sgS5t$%B4oG}slP$J
zqE&)Rmt$~#^acic;%LRRDb_23IF-dI2XsCCJIN;gKSkY%WEahS>ydRuo`(irS7BuO
zRKOIzD1h3Zbf4YrVXzC+8c@Up5&GlBA)1k?PX{@0dL^N+hpJ}0XV`
zuo+)y=V+eI=@4E5;OpvMp6cTjHtWd_PO-6=p5<@)3l769FhkK2RuDOEt%pjWm)hV3
zbzVq|LH+O61`j!MJB~ZpsGdN_hKSZb+_kP7vrA@&iVIqD3Yj-pj4?1WuX#Z1>lO9;
zjJp}vstt-Z#nVjqMc53d&F(f4XXF>Xtpk-0MeLn2PO^Wrut0kneTi
z*T)JzkWRz2^jzYq109jbUbsI?wvt)vtk*?^R%b8j(F(43CK)}KE>o9u1+x|X6CRnl
zlSaOI!h7e*1^J4i5&GJTFJ1Ckw^9t{&jVE2=n)ky#|prv6d)B18T06H(MnKsnD
zmo2xE4W^aQEL1TO-8L!JNl_y>e!o%
z!n!F}WEAIY8j`$(WvP;iZG*L(aM}Td#l2ibklZ1eX%hgA&5lufT%IuPq<^9F?}!bp
zoE1wFM{Jx!y^YQErFOWds)=J$^G9$WH8fuzf0BEGa2GYr$<~Q%_SV$-$rf#0{EO2g
zn|IE#>W73X{M}CVY%IRH|Lcc_gw-4B%pSG#IQ_(DjgJrny0+nRypE5aiZaih`SuK7
zVaV2~zM$-frOewaG~VnX(&pxnR^V>z7>z<7|H4q4eGMJ|6VPXzL4KVULy;FBo3C;=
zJr>r%&aO6rirwH66dtWq!BQEu5PYKB{gxOd>qae|bFI-)O+y{)uA3n_1Fzw+jnF{^
zNLQ?N**PN*s|lbD_e;G*xP9cDmmgFTjj6|CxXAajRKxH*0Nptjo?9
z;|~S+2^VV#q1svVJ!*Z49p&Ya`iRiBy2F=J!*KP&2n#8~@4c0@_C+LLcC*%>tKq3~
z!l)7mORY8OgZ2Y$dQRVLBzJt^XKc?j$~3~g@GM~^1||et)L;M965YBO)H!U&b!|mm
zSY{^wQb!2gpuAURD-+>gHK1J|&scUdW(K-kk3+J&+pM8ZZ)ie>}G^oPxM6Dr|
zJ>NKU`eS{p1CE&(R}yN_%d6q>WhX2+AVl6=8)zK`PW362!W>N~?f}lSdah_OE(SE$
z3M6r!E?4!JFadg@t#@)FIB=6JvpcU#Q)DEY$hO4$T&;S?HLAQCRfq?TSET95NF&vh
zABn{GZo*BY2e&&j0z%yOJ`e_3Y-qtd1!{&#)c9ZQA?jgvSGrWs!6_$!lfPK;9P`F}
zOFujsI%jOj_M+Q#ILJT}PE$<@iIff8U)y+Q%AR0e1L44p_n)xwt2VmQi`ke
z>_5}4+~Fv0WaMnV?gM?wE_qLG`!6Nf;|y+7b_WM7Xh%5|DqDzf+~Am&U1Yryp`41-
z);lFNxwt(vHqEFb3@)&p7W%xLm)0-76+)9Soch$csrn-9sa)^k&p9t!)2pxs5BCXh
zi*`2v%2lhUXu$2guWw?|%#sH7~r8EP_`4SNzw@kjBQ(OQp5MoWuhb=03Wr~EA}bYe298wfGoN=?#EC5X6O
z3=l4}cUYl01px!JQZhjJ#tCe=@AD*z3HKy*6{{#ggE|e524wd(Brefg)Cx)RRjQOm
z=3Raw&*WsFmEwc4RAO0M)nopAn^h(oEQjx~u@TyQUYE44)~ICTYPHu9YJ`^=)v9Wj
zKzO*#|4x_)v5T&0G+mQg+pSibu;!CNV%8|D*
zgZQ9!qqgCa;DqT+t%qF)YnND}i!S^rt*WdY&9kDU%&p#=yWVS&O-*>(P20#}rj>sp
zxv9{wAlVROJ~?k(AJg0Gm-ARIScZ|3gu7?RBU>r5t?1#Godw41<>l$=#915K$UwQI
z{YFQU3(%PH)y4@HS$4q&j1CDzl!FwMDk12ZXtz?`>EfT#+1GM~EX*+qsij=ZJLGmu
z{RPo+M{Yb3gqj*X#Id@(vpp!j4a^Cwn32#7lb4UGCu|(;BIq{sg@VsJsV!!Qb}y)teKc!obSeH$Op}t&^isH*z0O_w#O=4Ie+vq=DQTmOzj8c
zsql~FO(}0-2Zjf&QO77U7q-^Sz)TYQj?F6Sel#C((c_zZR)HnHLsXY${zwC($!K%3
z4_|5^Sax{w*%L+uY=7kDs4i*;I$1gaFofG#fJW@!YR>J^JQ)V&gpdHHpl6k_iY#KG
z^-H{v3^*S8*bBMBRgZ6w`q|gm?T)kP=%;-_agxL_r#Mt=ML
zQS)2QoOe@J<4|LYI9y;B&3Jt|qDXoRZe!yUjZ8kQZmlXDqMV4q^V$6opF^98GXhvH
zDO4cp>AYj`mXl6?eAo)twg`mptq1=If)%3LE}bm2))A39Xi?SYWb;W;1vvyul91R(
zIP0?tQ-BoFK;JZPCC0%wUCDlgttd)){H^_YDml+~yN$=(pB3?@iux}B*Q*kDs!(x%
z|F-lcr6ubv@;iff&v0+PX*J~x<3nf-vf2BN3(VxrYG83%ax<`}qHhN$YVs$bW1XCo
z?c-FjrpnNaFP_zQ)kIR$mHe6P(yUr)^*T0GQ6uZjy60VYs5R-OX(*$#Cvti-{l4No6Q#_Al$#42
zcV=VAQQC~F(r};Duk~rD
zg2tK>g*0FF->}QhVO6+c?&!3N_(gL7G(TEVC+ubW%f{viK>MkyexlcAdQ>1Raaq=gin$4Dh%kHdra{(i$Gh-xgLymI4Zqzy;fSB08
zM;csoEzX;lo6Q$%Yh5RYlWm^ADl1ovxvdEoO5Y}?MVX4O15xrgs1kJ9=hCOM65v{3
z0)AzZ0!GGN=Gn&!m-~Y-*T4X+`0!|@-sta3breO+%zt$dK
z9|dKxLoy!dLm{;36INQAB`jkU5O^#n3C_dF!9e6*z*nMnrf|6ss>o(ZXA3E)%exa3
z`sCh0{JkVWW?Yi-CjHH*doiOxaJ=oCTQM_C!$8c|wo{;&lbj0`B1e0HX|;3(>EdGvm_KRrZ|0l%0@v$yhZ*y?5W
ze59pVEEtAK+sq<~nsphlZ8U0D3e+BqNKd!d#l^1j%5^ud0nkm98z=-@>&5^0Du6L5
zGt;d3ND%t8ax#{VjpMI)YMfnbXK6|mB`{+7YV}eX2e~CnwLD?f-t3yOYr{3Mt(f>S
zfl-p1mQhS9JA1wvi}T@DrIzNiE^fk3>V&@(5Qc=Hq?a+Wfs!sxl`f|t^Hb#fPvJ_6
zb;$vF=}YDoB3(QkvCN_3)`?o+b$JuMYQR%C*#H>nd|+U^!6P+9E^k+$D)1#P1oz^4
z@a74qw-(ia^%mBaUVWodW{7zm!ADzW&Nn%7h7t6Zy8dQ(kOj@l5zcwcBM35
zvOwWiikdZfqLj^|TS8W5K+Dgx>u(KWJOht&&C9R;+H)q)-xdq~&2w%3Yn5r_=wkT$
zI8@8I!wNGo2kCH$rxMQY6jhL*2=H(r@k
zKp+5uKsBDzTdOEsRldCT<0$7}$n#iAh=8!C!sT%xy(l|1WWV9oE*lSXbs{h`0@nX0
zFMvtnu-NA&Fd$$eBnB|gDc^rzZ+3Wb~ysi}%YrX@F4
zRIGgu=!u{Q&Khm5i8+*=NHd-Q+oHzgD$K8X5B1Jp}W+_gtRq45JFo@lw9>
z8b>ZP%XaLIz+OX(V_6FD%dy3OkbCQ(#GivYox8YMtI+Ff{cz#+p8d_70NElYvHX)r
z@{a5nT~xWOH4e4~cUW$@+ZfwWa@f(uCu3P?h+p01=yKQLQ+gID2z~GP2P_u95xeXx
zX4Z#cl(7ILC~wFYB;j*3{OzLAX@>goQDm=5(ysZfjzAdWZnew)x8(*G;B?Mg*I@K#
zo6@9I>;R@3ARR!1#Lm>&3u8>DDPBzmN)_=@|0eX}_tK-`a1xY3jiAluR8*ViZ4YRp
zqaEa53*Rq!m*XVCwj5wDS4SoC8E{gK2
zTT^dWvpRz=`IoVoa
z&<+RjLTaL(;DGqp&#C7?jlL6<)m5`l^594=n$A|R{&hyc^^Tb^5*3J6>qh6fcnaNJ
z;RaZS$ZLT8S)5l%^JLGE^usB~@56Sl1oj(zP`RGB<+0ATk0z@27x(5BT^N1A&VB
zM>lhscMrvP|8HOIhxaYI{tqMmD}WD#Yy5$PHSlqIdEZ_Q2z3)@PSKVY%QYRGe-XV(
z^+DCmtGku6q1tL%T+L0zp-`GL!3(&pP?^VaR9ElnYP0Rlw4*wa#^*IZ+ZOvbKhW(~
zoHI{Z8E>q@Gf%&eF~~L@-Ryg*e2uzlJ_4!;G_<45(apneY!Fhu&hIC+lRz=3c&eWo!5HkmzGkbhG(Il$M@E^Il$X=-PC(<(fOJE
z)0BrAtWSuAs@ldGS$A5r$lKhM5&AdJ4pY7nB&N`~2)iC!ctJQ=eSEgd?XQ3s<(x!&
zD(`s;EiEgN`sX!|ANR8xp5bDoA!s2SwvC;k7`2-9YFB?~C>_iUy^*QDpm+Crq9}SQU8Pkt87#GUF*MfwQs)8U`*!ZxBq||
zA2DLOTB7)6?hmGC8vN9DXvDWF;8!|QEvh@1#<)Z)rNt}1M(^+4z?bI&bAbb-?iZ~b
zMHdf>M5pwx0Fv!6;anVjF(dQ2KI|s#mjj~R^B-w8J0Sthh;CYlZmGV^2EZ@bI(06a!W*1QleDzquCZR9d
zBOSOtr>i8xjqf-dH(NOfm_N4c_tGjfT^|i(>ng_CqzjZ?)qNZ8U%_}a*m!-DPg}fU
zo1~5`d+>*3O7jXhG_JS7W&0Px+Pa6}I%_aV`JEOMl+||%yk3qA*+s!~{G+Z?Tbd!e
zwf}t|-SPREmUQ6zFsElO`Nw}=UUW_MaX_qYmqY>o@ouPaSkQAMVCZ{cc#f%;F)BP`
zp!30GxLX9L@;aq&$@=4kHwQISgfDmdcKD2nU<=WL1cSl)@hd8u)=X7~4GopF%b)#tG!*Fb-!%<6QD1~8O*$-;(bO7_gW0h&!|@x<**
zUgAx=Sy0qi&
z%nY3!gN}Ww07;gbu9pVps>}7!*ADD+hQ#
zgVXkI0c?oi|I1f5AMHucE=ech`2=Lbc9N%pciZ?|s~4IYarH;99YX;b)xvqeR78^=
zS3g67Ui(zYbvF&+x*YXb>FWKk6R;c>mJTG-z3>$Ugs*KYAOmPmTF!%8Hc@;<+J0%*
zyCk{BM-jrF=SObJBxK$v5zNQr1E~ygVo+lT2gKc_`rRSoPWsexWnkSJ75U}v3fEPi
zpQ&MKNuoM_aN1SDS-f|(nf7j5_?9&;^nqQJ=ibclp65w~L^Y={v1{lTKqxaR3!fMA
z$a~HC2JfRyX%@`CdAumT32;ULz!~yD`q%wa@6YxBuS)-Sit+zSC;z`*9#qWox3g&D9UZOc
z*y8~<1Q)3PeZzUY>`G=%|5mWRX)ZNfS!!n$tdPSfyaJYOrwl;FXb1LsKJK#>Kg6J??@%*=wJRM~3cJ{DThoQG*Khxy
z)qd!=p6wB1^6!r}&>-*rx#u~yy{k9wdwhK!a6^`@ei_^To%OGOK7R9d&&PSo(zQ}O8Rd%In;7*AYt-~anrr}+1nx2)^`
zm#R;zc(t|rSb5w=Da{kVkK0$iI{f9J{l6C$zwNCqTw|ZFaQQam6SeRc^~ui^*)%R9t=R92XzWUS}f$kAHp#H<8mRa1aJ7R14`FXLs%iiAkX?_34vBm2DqHMo?l;xUU`D*KrEa3bT
z&;_ZWwq?v|^)tOYiqej@o>cF
zJ3(Nawuj%o`|wZDL~+|qyE-1v6JPHmsrSaC(DwcAUDqBzd=CsS@#;BtJHj=WUWqdU
z?LNuo1GO%%mndddym@(gaU1tp?(3Pq9>mA}-_q&+|I1NeC!>A--^Z(6IpsIMRs*fo
z1SP40-01KB&vvg4JbZ6C|Gs^{me&9Jx?DfD6L`4v)tE;gHrN08@>-m!=_M!=t=5yO
zKe=qm?Xt4COPO)Nz})wxmfwE&+ePMLGlALT-qopc-}d}iuJZsCadM)^3)i0qCVbPk
zUzdNk`*ulm;~`M8d-nUhUGhnH?a5g4
z=H}-s-RtLCe0)~^-G0~q^Yeu!p1xL7@nyGqTy@rY*Y+J3_HHQsKfCaGg1GHn%NQTf
yQmAWMR_i(i_`((R;Z18mTb6Mw<&;$VF_;g1A

literal 9719
zcmeHNdsGuw8jrFnh#1h?XhDNo#FpCB_E021QtE@M1?+BJq)Os$HI%w0^3V*AWURcj
ztbwX+stDP(>(bo_siHh<V^$#MZR3n6iOjBq3rVnGo}S%-)b{A7{_m-9Ps9oSJ*S
z$V?aVn7($QOeoU`HCP)1*Tx{X)d69&`HHel2?T4yFc~jh
z#$(EaKzK56fe%t6(_k_TQ`V7{ZFN3HU%`|ZrV%nVLY_tjM5vJgEljW119X4_v@vBH
z2J|#SOe4cIpwR~pV8C@4h9#}pn+7}=vhS7bLL6>!weL5rR%_ag!$spZu1kF#I(=`*
zQ+~#ltuQ@xD35dSc-^{i;gQbiYPY$^fjYFM<*AtUw-P)4N_5|TxU5?;p&8HnK4;hrhg!A&`1S8!
zN|Y-3kFr9B67??csLNK&k=8FqE_@I;2sq)F8wrJYX??x|4^32esg@!$vyl78K)|;^
zg%$jp&+AX61`YyF2t>l8Gz$(?*k(yMWU<3i+k$W8pm&O#lR;{2_{sfWMI(ukQdd(`
z9E+=)yk)kv%vCw<1GA*pR{K-MT{Wkm#H`fmGKut9OHY3iF=q*49*@IU@Za2ZGrk-h
zJJI7Zn8rQVY6*l31R}Mx==~A-NEhPeINm`E>b|_h2v?mhfT*x>Z0vG%6RS}G7nIoU
z-a4K|3?|LGdUnrD^;b_-+d73P7veZdy_)A+Nz`!KTwl(T*0-)gvTWgvDczm#p@W~W
zDICXTi_w1fsKCW?Y+`#3>bSDAp>0k_PwD=+RH!redHtbg?gUEdEFlku7IXBYeBxP2
zeBo2@R6Yd8>=Q1{{S72YE-_C&J|QpajYBlMr@CPXF~%m41mU!alYX05sXhJu_l_5)
z2;XtVw6-j&NXT6a2{=%R%i4DDUTYDn@sjI2@M82sFRDXD)2)5+6MR2F2}CV64HFtu
zC}NL$@gf>G8i{&x3aBYcRiCi9tEkv)9j`H_T784D&<~?UDT%ApDX{Z$uc(__xYhN8
zfzqNkcY7~SIH$=y0WK5ZaRD;>h!FXolhARiP_Jk^yc9`(%%7k{i&PTFa4X3z6$F#;
z>lisN1hcN|D{UYlDPVeZlE-LE4*PSWVz#lCt1}#JZeH*p
z#hc$ydz5rnpD~n#f6WiKr;2UfoAWTWiWaDd9$p-6$zL9VIBeIs>wkDV0w2M4K*a~Y
zyMG))b+R$L!4QYtN?^L@XU2|xB%4}miks0EkpP`OdkJp}ElPdn;saRLZ^wv+79u)?
z_xDx0dC&jv;PG`=yPsSIIQcV&k8w7>5lm{%4v%piyAV!$qcwVyH70KND}7AMc_cdmNY;01Ei%=Z>SAbhXDVX0x~!LwQ*?=mpFznf(U1CQ-<8j%0{mPIBBvw|SAo_fVgr>^R-)_ATJnAK
za)fHXIsPIdNV#xQJM8Id?Q?z>GB?y%-DDF{D4ijnp#{J4U!7KI
z8yv0QJ_nFG@m1GA4SMeeVyf~*KWQ#
z;e71Ek3VNdEl~9c1`b{ci=TO-a2v!DNJ{@!;;ZBG?){>dJ{TJ{_
zb*LPjH#s0uSu#H_XS`G$q1uqlT)FyBJ;AKyOA<~j{Q7ql?Gd$+o95hkPO3b7B`juU
z(L(j!FGj281(E-VzIZ!-?U8AVH$aiVvtJyX`yang2MTAf;A
zrAzV^ROt?>D>@~yp@M%ZwuxBK|ou!^pkF-2gcjg%}slQ_UJbZYi+sfE*|N
zr+4md=o$EA47|cN)zCfd0#%i5Tx*iJt(Nu_&#iC9?{6m#kJt8bEC!S1aO3H2hbOJ6
zU1C;SB-f~&E$8l4gcyVFCnInPdDuJoGd>Gg?68M5y5Xt52{gInq5z
zdBD7!k+&i6&Kv(Mr3|}2#%__8nzlI=1cHPL73G(ij`Sx`QebOt?EI#_PjR?kW&8F5
zv`5?j+QN!Y?mRoudxg-mCn?_tRBanOSMZI}CX%`wrTkuwca^alZAM3#J*6q99%yk)
zhwU)t8iDmi^Hnv|B_3UZ0Jhp!NP(TYet}c5vqOo}jH8w4KQkIp$}0(Kjp
z_WPCXSdX?`>wH|b029kH&F=!GQ$Ptcy6Mp@?@furtlS=^(~V$+EIgl8R#IGly}UZB
zGt6k&TS)*tF7_E(X}c}e4A<8vZ31eoZ4h?d6A~`2vX!c1#mkVk<0}ySjja%iNF)$e
z8YTAW@8-e>O!V5U_$q#B9<(<$6O3UDlG!5KH{!InrYTHfC%jBBG}P54uAU`~M-=j1
z9d6O5h03e*!QfVI4DpX6|no0KIkVT}pocxrVjD
z#_}T9^mpTo?!f_0CxO;!wQ4Jk$Lq^kLne7&ie+#h?k4c2=k=;~ugkLgGv8R37nzIB
z6U+2P&H)xZ=h6_8)uY{P5UIM1H3${!I*|cWU@rm0g>=NN8pz^~*UI7h)xh!s^HDyb3$x+h=Ldz58ip!-@A3@%
zmd33IUK_e59J+?ms7kdpE`e^>iyFlFo+OEG;<+oJFP$Mq!UFM1y(n33aD>VeTwRCCDc#b;(~UA#D2
zybbWa?|MHBaQ*((2>iu20_{RP&}X4BeqkYg;jYmJt-10&id#pwq(rNdJ5-EZhMdfo9?E4j3BfdVk0L
z(fxhExiE<|E`TJDKdKnIp`p=6+|575xeCq986pnbJr0HL9kQbNdqWx_&WqJ7E;`k&MPLzDl1
z7{b)bFF)}7_SsO~#G+-p!d@9#1wMq_{qweWV?iJ{7s!NV*D0*{040r
z3x?a;du-iU0iOF$=UCFc$EaYRPhhzj;IkU^e}Av0o6yK);PNl*bhezF#hUF8pB-Pa
zE$^{H$OY+uPgZ$B%*T(J^G$qb{Un&fA)PH>!rGP?+AOwce`+Up@?$^UpnfuQ>)NG+
zh6=WE%W-MEv+>fr=zx#be+RD{oH8~v(k}J(a7kVLSkgF1X+kr%W&n-74^RJ08>Z&=
zs^UE}BANUh{JZ5!C|H@2p82z%8%B*F0pI%<<8gDR5`4aIlW@D@Crey+)PaH0gV(L6^ymoXISKzOmS0z
zv8_QVt^%2n=u)CeL?n04{uXA!|0)l_Z^erI9H{;Rw~#lPqW!_1-^}!pwyR$1vVpu>EJZ5WWLfA}O;Oa7B(s~>2ft%9(7B6h~OtqV4aTAL4n|+xA;et-faFIrAOvtE4}_K5$?Two=x1xtmH#9XLnPOR-yJ=Z
z6kq}6UD9KLeqeke1*qtjuzYQ*vGU4i29i3nN;`Co&9Y5{8ua&-bj{)d)wIAr@i{c}WV!>)nFiH*RNCg6Jz
zHt<6nIIS%Yjl(BIu6YVyT@e;f|DgJ(Yx(o?=(ghwS%v=`Azp@(zj*?~t?Ee(0xb*A
z{zkNmlVzUCM6i8CY4O2`uv}W%;6uts!)C{PtNQOKVbizdNF%#0dA`MfyRG$Z?6Vu1
zuOYI~%#mzUkG9Gpj-K6=os6R9$bkkk4WY1ozPHdLuH%4W>-OF%)vHF`3d#Ot<5_1D
zzJaoHm90)=KL#}GAh1Z5?#5WR9wWT3P&5{lM@`OG_WEe$=^Lo_fRJHXX3kMtgFp*D
z8Z{cMZBB(H3VAW3-1sbLx^Pxp%qt8(ACW9%{iI`#zml|?Y9a2kqS7bs@EV?XWHYwE
zkUvKdRzB2d!e|h$f&@~_$+h2S%tr;|#Z^SB85g}Uu6mOX9L7u_@9DEUE0OeJ-u;s|
z*paXAgwY((2#`t&xIbcIt_lfb*_Wc`(!g_dRS(FK9PN=qcr~%i#7z&(F+PDss4^lL
z-0hg_w1Q9LAz*Kx5aTru<)P$JaJh=hWSkhg>%Y?{5hC2p{*GQbE4m4MP;Bx@V}?t>
zF%xZsnz;-nS_}3D4uqp&<=+l%d`H!-DIxuC$dv^IMz!AW<-V8I9gC}IUjaf;i^JC~Z5B$s3xx?`
zj>g=!59cxkYVs0bt7tkT5U-@jYTZ@iBiQboD+wu|rh!0Ku`_YkZskVr050rmxwY0B
zIm&m_A|dvtiMai`tFkR^IYU$W-JuH_I}WUE;O_^%?r}tNddmim#2PwRlt^=M%PBhQ
zwkt!|B*-*J>+WJWszF8BkW4TDm^~8>_dQd{5h0yD*+yghO4scJ9
z%+g4e=hGXybnfx5IoiCjIVSxS!6Vg2Bm4>B$NvGgOcPq$Kp56D6j9zCvmm4oJ?xlp
zmiw$h^g)MCD3!j`y%V+gj{1X&%oDVi0G-mKETJx_xVI#H_4Nu7okNqiM7D_$*sL12(Uf@Gh}`R~X+`ap(ySh0-_ohLMLPt4?7j9(DEAj#=j@YjC{a-<8J{7$*IWJE
z=Taoqw4392b$+OM1B}@J(o(j8_(?0CJz1oFmW^4I;b+c^rX(?6K-2l^g-We2tn;)yf6y^5@zxnvvOP((wwe#T#d
z0Xy?8f^G&ElE8nU3r=ydrfyw-V-AVbe`bNQL+ZXV92CD}BywN)?Y=
zq(u&Rod#qCogacaU>N@{ZX6T(Vb_9fe3BLL?@-~ZGd(PG4LL`LIH@hf7j)CEuUGUu
z2$XZj(e3{wAz&I5J}#Tf5{u9X=2T!*hQ65o8+h)XL4)-=!Dq@N)Haaf+=K3aPyHrL
zdES2_P4|gD{T6pKf97V?j*i3P5fg^%LwkwJvPNmp>$OVLKs=iN;8I+WVs@+r!{J{#
zlKU(=^--@XGBay;u}Tt^F-0f$bclO_%-d-~JvrN#YL7K@{683igE!7QLt
zmWCOm2$oxyg`&ZJwII)h*7ufA;kMS7&r`Xbq82T7Q0Xj_D)v
zDHaof>G48pW}mfFW}s0QBky5S=2ZZ?4kH0yDdEJJepFN0lr2ml#ru0*AK%&)3bf4C
zE01x_vKVFC|Rxuo(haaR^_?_6;1Cb3^soHcN+oT>m
zQ5Ou@B?AD7ZOIG$8Ux-`fGhP<4MsJIyE6`o#`;zNNCa$Q;qNZ7;LOXDi(43%@j%IY
zD+4U<$1q$w41{TDvjh61B@!?Kfg#=!*!6Gj%w?6d86r^vg9CleQ_HxE%)qtJLDxS`
zy4L284}bd+G8N*oef07)hkreNY|hr6lrq8(Fo03wiJ*EQLQd{#a&Jtk#~xic^_*I6&08_@
z2wHD%EbNWM$0$}h`KU8`D>FXlZw{9yt0kl(;-?yb+pGp=5P_k7#3LJ!h2wV2xb5ruVI>O1L&D
z&Ej%rYYS*HodSPHM9LEw6$}XAK=eVAT4-9>PR%rCEh&DFSm4-ZiY
z12}F=Ap$UJhc#p##X>LY&ify|D$q)OW#PIfx?;Ah?5UG32|y8XosS^6u{VJZM8qRJ
zM5t|W$?{|O1!;nbmjLN;l(+j5${v2aAaipn{XArFk}~Wabf%g<;dP{A0YXzIl%$$^
zHI57-(&|ou*eAMItO2(bl%a`TrSGiXB#RfAi&I)(zA%Gs-;b{5c@{;a@mmg>4YreE8BAw{n3?kXF0fG3_NK=nx|^S3&0U^riv%=e8_0bJUz
zWf*D1dKx=n4hcq2{ybNrJ_HA!-N<+
zyzPDLuftRZg9Blm!PRm3X6G{;AeX;H>7HLT@0S=$O~3fk;(|&{DB6q4
zihPm*WUBti9SA;5M&$UvvY06{x;v-hA+v*I>-s$iD{MNu-X9JgXxM>8M}Dum=P#X?
zuCTR)6zlCBs{q*aqvzt)*7a2HY)OV!hkdS`Phive72(GUU+{%wp{9V@-oSXw+}n(i
ziT+l6dS#QHBJ8C=%UV#GfFEV+qLHvx(a9*;I;xmH|9FjxjKqrufFu0&2@m&o0pkO%
zhEAot74)cn$^fT*29j#N@!!D|&bR1M2t+p0$%0DkPO`fk%g|E)>irLjEEFLHRAogA~B5Pqu%s
zkdnJWi@~8=JyY`e6SC~CvpqZ_Lwvf5A62o2B{+HX%@JI@aB#JSmAL!k`$eI!l$L&V
zN>!rvdyT0RaW~^+-e~Bwn+v1-UOn2q00FnF)8Q?8oKO+Cm~>6c5b%8j7FqC{e${S%
z`8%w-mV>te?_JHg$=}#>L!v6@?RKQZv6G!V)5TZg8h^w%eZnR4n)o)l
z-yyO)=VyEoFUvaqI;ic-pGKecPzOTB`QeY!9CiVLVaMhfuVfJOzsU%njp8s>W=$AZ+7SvpI)cPQm3l_uX6
zf8;)k^$s7j&UHCf8n3kFmRKNIhdYB{#`@6+1u#d2JGYB7l(}hw8MW(j}_Y{Zn
zZ{ggZeNN^W3NE+)%A?NNqrB`j2D%?XcTa*3X3Wv*z)y^PakDKOa*vb2D2RFP1tJ@C
z6Zq}S`6&+>k-;(%Du3Mfu$(+<(lB0d%|t~3;0u|KR$4Z80meBR3|EcO&E+Kr
ze0xfF!%m2V1Ua^L@a+re7kSTfT=aGw4A=q-Fy|8^vR9e)HQ=6>_S`oWufCkTjCzAt
z&p?o5PYdj}EG}VVniXGbz}UCJt9oxSpAoy4L}%iVidx}sl1WV%Oh=|pv7*xg0fM3y
zA5Nuq-$qw;UDW)xPfwJ<$r5Sxz`REn$WLO6>Rxu8!{7HjE;xwNA2Fv}oEK4f`>QMQ
zj}}1c`A=ZnoU&rNtGn9F6L?2ptBDxn*n0ptYR_pEej~@9?~|GE${}9fQOh2CGcF2W
z^DF_zm)=P?;gX%`k;u&9M>_FMP>={xK$@;%sFL`(7BcK3V)Q-CN_r#$1G9WWijTjZ
z$U>JMLz9ZhsHzNg2(2XWcOdEsJ7)%2T-7093;?GI3%xmb~2o?4x<^cC(L%*qiy!SI3cyO!*wUQ!AP#TU}PFrv+OCY?lm$Ct=ixrwvB&yyv;B*;wUYRDcj
zzz-&#p>{lRUO2TN$ltTTk*7Y`N*{kYPvACm6Evleu!?s0l#CNxLXiEki99UN-_MkptC~Y>~b7n0#)!l
zhzrmyFm0T}+h+use*(8@cf*WH>we5J?}%jc%Uy4%PIWxRd=mM61FuIShR&sFH`kjg
zs|A9(O1_#1!omP1+kkl0+*FDCQt(H$b`X>D=u{`3Yc@i5X(1P7yQ|IUuOmun4}+>d
z6Cr5Jx@sVJWahpD!DXFgfmD
zi4~3H6t{o8PXaY}1-X9Ij!gS1pl(GmO7oJ8r7
z1VL1doWEf`T#e%Cvt;(KBo~C!%1-heBS%EB+5F;q_($H
zGW?^k#w78=jy@ETjB(vltxWS)4cb^lw8aZhg7UY^!Se#SUpi2eC(M;(+Z1kpS=vIY
z=5BIdg+5i)5ENdMa!rV%n22t_h%}bD)Llc_*S60DDtN3ZzpQ}lt5aNP>mmpO``AgU
zoo^|Us;lX(S$d+Ch5>!C;IFiodG`Zv0)a>#k$i@&Zu=JoVQdfecqp}T0K^rTv|qH`
zHeJa1%0+#lBWZ29_O~;bISdYqE87}=*{nCG&fyK|?Di)%M`(s?70)1ZX&0=T>JTp_
zn|1-w>kH^gF$j4EQ}?%F#>Xh8Xfr(35nwgUY9O=e-HGNzjkox{dn4V0+;P7^G|+Q+
zM9zFxBIx&NJVa#EW1&200m=UBJS@DHMo|4XwKgPD&|tz2YmMlw)C-dg{mm?ittaGft&1wAnJCn6_q7kQECqD50*NAN5zIcuqg
z*76e8acq8xxJn#zJ{G5Owki#>nmRen>1^kaXki5jP~EuG0F$S=l%LtFLABnWLK0+x
znNw#i$jQE#73S$#Pai$0i{a&W`#ca}?CKh8EA;HCp@pkv+;872XAg9P?TYZIdKp>b
zAaL6#L98li=gHbbkA1?|qd(`!t){A2=N*kIEC{4$?
z&18BH*$Z=21;4^}ASc{k$HcrkhL#p>EKoZlR!IJ+HFr_~A4OOuxn0e2?9v0ct8g?Q
zTuk3~@)sd%z6~s-(_bdwk0O)QD$9#z4{E_l&d+a}zFdJDdoNS)p4J0a2ut^+iF1X<
zTVZl?Bp-X?gbSeaR(rs)FOv{L%6~5!gAv)0>rdKLbu~sC6gQ;j8=o+VH#{}BJ!$^7
zh)y7sZaY4-X!K+`28yl~Bra|h`<)|^8F%Gb)LL3!Qg+a$_74v~mgVR~1g&C+g)B7t
zUYPR0%!c^SA~kG3Ki1r9=7Md^ts=5NEGfQmWW6o|FX7MqY^JQ;WhaQcVyaG>@X6mE
zUCY9&i(Ka9#4S@sd0R6&k9eK5sP!EN6DN%tYe8#UG$2jJ0AI4PeV%bFM(BCBzU%ai
z&u#I52Qu64B_T8`iwV5EQM<(T8p9?6`cO%|>DZukg&S(U*RxweR-VaHRu}SHWe{+r
zB*?-~ux0NBAe{q-ExkmKvxwIyythUUH<0_1Qf;u~oxP?#-aN21rE2}N6v)!((;CD{
zarJD78tPVU?|qEw&Blk5nsY`dgE7+Z4K&Bfv@|5j&%DaWZ#^T{<-pIHRqR9J^%8a%;yQr_Mu_`FRjZ*1^|W3o+Zl?v>1
zG(i&d*wOPtj=7o-hjXCIvM5fJ*X_?5*UEdTv|a1i8z1nwya7Eq=r_AE0UNi$!npTW
z#_5J&>gzOSlR}+QqN5P60O9RKGE!&B$(47|cexHoEpkV2fI#~~E}s+fcG1kUMBMIp
z`6andjV0LE%T!BKDEn!u<5~!l*gtyZ*d%sK)6-vZPxOVc$0d);Pg>(Fd*Ne6tTgfS
zF{kw*Iv>7{$De7fYgIyoqX?skH6EkFCmc_7ELU4O0CuGZ*!7X^^;U=GkqP>6$lbiP
zf^6d}{T~azX_{jtA{Cg{EeE4T__7^_@s1znUkFi27*15yzH2@L5LE6pt|T=zefjF;
z?w@^H;%f3E()e%9?YYOoeC2<5L83GHXHT+z{C#-=jpEv1ZClOQBH|tR;MN9D*VQzt)_EvF
z0uJvr?G-0~d~K}3YGz{q5isXGY2;McenIY1bJCxP80JWpB;uqg1u|R@T95<1DBlqZ`exAWJ<$&eZNSvE{Z>d
z;WC}`uzi!qGxyc@s)u(DPPOzld_XDTeukmdgP1%($f)6I(h=)
zsbF7zZOH;iIuM~6F<%PYn0T6BaYj8)1_(#l=>ghzatQgyaG}6F^0jD@lzuGPavrpPJC9}B=aKfNt_p~5W|>uC0DUc$gD&URBpe^p5G#3`1nSW8
zhKzEJ1>9eG2^HiBs``p1o#dAJUGkeqI{rNtBqxv=7UXSR=2RssPGPF>I?
zkO|7iLVbhPM3~tWGp}hUSpJAv;~niOe>>I%T>_$UQR1-j6q{pEDKBin>|F#}yH@Mn
zfsr2ZtEDg<_PW=yP+8q*u*YvcLy5zPD~j#}pN
z1HJa$$O&(oit_U5_$BL}2Ufl```NH%pq;GsF;Et&xp{?xGjVcIY(<}`oqk-LKO6d~
zwM-vPJ2!3VOgnep;DC4G9R$|AYQ&-Qe#}XVEHcCHI2M5MdyJ{h3^C^41MDb{isG0`
z)&7GJs^1AP%s}zC@MY7uIfg!QI)d4&2P};unmK#g9H{eOZf@KB(tuWJ5^X&saGq!(
zN9EM-w(+$Tm!IjRD;NOSsfGw@LDGl2d
zF?+uHv(5Bwx}-M3FN>&J!!F*^R4l(EqLc+l%RYcDxEELF@F2WN&EzA&*21rmumLlsYFW%i%Uz0~nTwFi%YprZYi-?nMA
z)0yF2?A=}xZuUTzXHQkp!KClC)5hy#N}G{7EodRL06;E29fMA+$L-A2aqWO5cYT3N
zt-^oq@e7*+b&H#;ota}_ZIVDoeD%iBD$#j2%*)AdMpiN@?&aB}K!THk8Nv`B<045S
zuG+d-+qM&DthDjwoI%;=!Zp$sVp->DT66XYa5I3C2$Jb0c@B-tvXpKF9tW1lvEJGy
z>QzTnL9t}SPs)ur%FircS-c*jZqJB)SQ05yPj0iw_%g*LKQMMO52_vTEX@ZGPU
zGkwAzVtBL?BG(V{iDef6p8pvoqG8(4M3?QCyvtHZp#*;r)JhIdlVknz>S+R;1$Kqq
z>H>?~l^P^M+%S*Q)nb41cGl&bS8IUUiXAih=m^Iyiek{|Rh+Dx7*MnzE9WE(f
z#axu2q2mJqG)Gc>Ti*oK`&h&}N~_&J&HJ@XNFn%S2Gs)A)OGI&Ht;@pV;1^M&Lr`B
zV_4fHX7Tbbq{Od$4pauN+HURc9C-=d-B`P&)>p4%jyH`oZA!g}p)A@5hj;=piXXTR^_nNS&<%6Qd80b*qT9h$qCr`RZ(Fktry
zl~$MDG||gIoVVHNWk99<-QOCZLfqNh;;GDf@`;5Gbk#ROk{{kczrphc5~*-*3aO$n
znN>DhvP=V>xO{vs{yDV+%`7?e3%L7rcv)Yi>j}+!4gzO;ig{!=h9w^5aV7h22+!9Q
zvy3}$vlRHrgvUO^s0=g9Yg6PHL_bAbL+pUpq1Bu>vwn!&8NBtJ1$6nNEF!~CbOeFG
zhYOxbF(4^_6r-aWv1}sPn@~A_FR08De=sG(e
zd{O|WI+D-So+ls+bqopxscKB|CDlk4Du#f@|6WYCrxgabLmKj>ZM77*+k{
z)t~{Te`Y4*+xZdrueT1gq3pWTlnn63mAUB{6;sSiQbzLZ%L`(Sei|Ns&t93LI>3|z
zi?@(P#?C!Um1a_|4t?&d{0Y2(u1Q-{Z*Pq=&i!eAkt;WT*6;bp@xnH%
zYlce=g{0PU<|lI&lqk4z%nY=u-PT#gU%*Q(Qsj2Hz%Ry2xP|0wQThVx=2noTA8>w-
zl7H(c>(Q@;hHAv*1l7H%Fp&#V>kZAkS{0wp{?g6(+UiW5`9;p+AX2KGpzRd2xKUw}RM2^!W-+q{O@DZH;&53P
zc<9|A31VybJ0>TFx}@wFNVeaeT!z|0h*PBK6JtK$dA9*BE0v005I1SL2K3~kswJsB
zV;8y0ewp6eD#Ef0%GT{DNiCSn;S|n)2Bck
zD&=tR?iH2jpS8N?FsfA~OAPxv{qEydc>pl`NeR!;hqf_h=j#O}uoTF!Z<;6OH0q#k
zt>L)J9Bp{U^9up;OUhKk>o|TC=hebYUNK_>^Ta3tkp4zvfS-QcyxmK#)RNB8-TY2c
zZ(?bqZ?^U1v#V9J1~yKoVo_6L!LPLpW6FFZTPs+};2qzVD$^$>weqsJq#N|CH+Z_o
z;>5kz^fdPg59er#RvE)q^E}gp@OYSOY>{u3OBrhFQTxm~jM{Waw*GPv2}*HOdPg!~
z3t{3VNjFi5X;KFy;wLHll@?WGQ#-AG>(KKVuH6?|^Q6UgwH>r2*7{Z6BJ;O3)B!Kj
zFavYERUJf6@La;v?|6Evv=!ef0Gw^ybIyIvwd~a9WeZJOkghD%B6s+U&R=GlKx63j
z9{KXZ2!Ar&nD5*@lk$*CecYPu+M9TkKpDP@J?^~4NBMZWNuuT<34!y%dEmy=L$!w@
zXCU|0=*^RKleFJ5D9{Lykc*$2J{s-fK$-sTc}(2Us9@q3xnpI+G4R>!m-LlmAa}@~
zukc8@nR4Tqw%l
zq^?k}Me@%-0kO}NSC!xr@;0lMZmH-qE|Uch26*w&4cNK*PUJ=*D}cDyw&8Ud7#TM!
zUVlZ3T;hF)14J|&#TySSTNyQ&J1AzfiV(Ec_(aI-WJl*UnQg2GC}CQNNLMS&7YVZy
z`V{x?v0Q3TTl!so=EHx@I4R$N@<7OuAM@rraW43_dJI?sfRt+RJh|U$X?`=>So2Y6aq?!P|6sqZr}qO2G=$l0EQ4kLY2Eb?JfOzkBYeMF6q^hvyi|H2
z?Kad1Sy17#*&{e)XKs#d<8{d$VL+4~0H(?|9P
zRqf)6O?pP8&T8CXDgR6?Aks$e%D*5j~(D5?U(E;m+Y&K;5D{1
zXk7&n5JC}b+KSC1``db(-(IrAG#Iak9kDu5n)VS$UO3zTZ433H5aSl&ZU0_v{HUyi
zN5q63&i*I%7Y2^Q};{
zFcUi(rd|$@i=m64u5
zFZzKqc!>C3e>Bil@)^XJZ`|UjF<1<@cd(b&}E5hFy+nUJ&Yeds1fRZb8M_
zW@ct}eyL(A@naF7YV?=E6knv{1prqExev9p4%8qb{(MZJ{ZaC!;zQo4
zZ>2TmJ1gTjA{OZ&ujBI~N;!4^!Hp+L(aJj^t7CLmD?<*~Jk@l3Ee9Dr6plhXegOF^*rQHmT9Dp!X?NAnm<&-UOKkxYf4`YOI)r@
zT9|{tK@HcWx({+5eDv7!qRbh@1uET}1X3@Y8&`i=PPu&1pBZ?8@086y?seh-e_+*R(u=J)iyN^twr
zqoHqmDEBsj5vM4Xo&>I^2;Bn_RLg}*fzoonI4e_6I6DKyv((z|f20dCNJh6$BJ}${
zL+yF{wb?tl{P`d=UwybcCuC?xqVznl9f4?sr2B}p7G@(=uJ?>d0Wf4;KriBMyQ$z`SmV7CJRF=mNNHyKjjuy!DOcT0k6J6NoLm;dUs9&7Ee0v51O
zJL20DSG&&&Q;O>IfD#dzZq-Z(Qq_7SG#6(KZ`mjclKQ;;HYPN@3NXdO8zN`4`%R^C
zR;)7!0S+sNfU%FsdMp?A$@sPy01qu~aj~MH`?={ZS6)LGpqjE6=MCMfgAr24ft*0n
zl?n^ihI;rj>CdNRn8ezuQ9A89EVG#$hAWH>@mfVG=XtdyLuSO@u)I)Ih8~0XQQwI0
zuFU$o(!a?nZBEbXCYEtumJ&<<@^(Ap#CW#*{~UMt3>yfRb
z2-EmbYwYOP$jQufn=!&;fsKV&U)e$qIaLt$yYnm=wmX#y{;S`{wFwFWhm|OrJ)Uzy
zj4gJ5MujU!o+(*XXzr6yuTIItliAF7CHRsCPB^d$pZWyjpB}A<5FmRoP>qi+Qn@N+
zvEP?x&KM^_d3L{YbnzL-2k|!g{tY|M)>s}QuRb@uvx{~QL&VYu)gT9S?$)AbG#j@|
z74|?EzxdjD(4{O*)lE(0=5Jq}Nya&!=dW_10>2MvKH2Sm|KJQ0ut+AKyY}6raYR(lv-&&NVzGX$OM|<^41$?{PjTDiLQki@-Og-Hz!xTN=D-c(cvi
z*;>2&8L1kT3suA=TTiO1nux1`nj&_REjm9`kbj1|oNl_~ILI4%Ug2q?`!t@h~~
z$o|#?TtJ4km14yHidVz9SnYe}^sH8G6MHxLH&D247_3~rhLw}Bedb-&NQHOEmRwEV
zBt+!>cmiW~1+e2e22s52CHD#*Rv$A*DV6gpvdusXw$|pyY+NsZ$1SiVO5T1u#IQd6
z^Xnned8a%)Jd7Xyew6SZeF)s^89!I55YZ#ChOTKP6snxCDZ)Gl0dKa!a9)rLK0Qc|
z)?%8N@%!6k{wu4=Lieo#bcpUBb=$atH-tk6>CzSH;R2-_$CiVf+2Q@VP7lHi@dyc#
z*F|Tb`t^?6_Cw0C&m9+<;Z>qxFT5N=xYQE3Ko_yacc;W;I_~6vUD!;4q&2-1Zoy7P|$@qF4+t05wQqb+FKI^7O+&4PGo6-%am+Fn0&A;@&S^_xMXH
zo+Ce9^0_8s;z5bud*fVKZR2+i<(kyf6Mt>H=fCTvD8A|!C0$#qCXcNRsdzx+kKvPN
zQ4kT=FB~ERo#YVWQn2z~cC?9gSGS>_^X|5Ir-@-r*PAX4J+=wK3I_=TYybqEpHeLh
z4c&T67-B20j?VPoh1j1x{6!ud$estihZZS
zCz=TwJADOsF7#^}xVr!6eUVC-S_cjPUA`cGGtA4EzPhY#88V7T8~~-)3HKYPjQ3=Nb}bVDz#I
zaw9NaQ+}RR&9R1YX~*|;@p;Z+_L#K)qL`geF1a-|(avH_<%@n4nzn5wr6D$L_PYmL
z>wqAhUg!M#t9XVUnCDBQ36eDKZy4hEumw)VM^j-Yn}J^!Sf8oBrVXbY2%Js#ral?F
z5cEhPM8>n>F0Afn82f$`RnTwnV0o|a(RchVl{;rJh|c4*;&I&
z8HZF3L0RGP3W0V*QtUv`%TaH>A^$&&<2qz2-@d{%L&RV{G!P!
zI>Wz~@k0g^qUPNimn$X3L%@vie}HpEurSXm+*Odm)DOW|J=1|KLH}NA3`h3)KE4c$
z=K3BFE~i25ObWbQ4Aj@|p{jFm&c%K(Nm{>t&sebkMVlBQv`*2;
zPlZ_oXk)
zL!QEr1j^E^9dK$xg3
z9wb;2vlp)~uEiJRhOurSUv(ZtF(UmZUlQaRF}F7!qy1s+C+{6pCizRu*Zvt7@ftrp_-23)+f{;6$TIfeM
z3f;|6s&~;bj~&vp?CWLo%L;RqCciR!V*2(oq2gX}Djgk+ReE(U-&u{!&>I%>$J0Gj
z?y>LSQC$5~&M&qsq;RG?Eg|=;07!&Y_QpQYJJ$94+-W29u>$NCwf48(AR)`&H`w_Q
z0=90V*7f)Fq1=AB>FzwDtF1-7)I38h#E#q(`^r2oiFz{IofhV2%jS*9^6UZS5hs3Zvq3*P>XOtKi4RxoRTYH_&klS!wZ+~%5uR{th|+JM4DhNn
z((UJ?0SX=g*DS@YNnP#5V4R3llhP0Eqt@xlK&OER65o}>N&SXhY9yxu~&8^)uugiu-%l&p!AbGy!R4aL&s;h1b}?#
zeC?DHQ)S}D^=y22qOf+ZhYT`f{KB@FdCunv(ck!s_J~uq8Id`%wE+_;(ADHrFL*zc
zj&(*H2k?axrv{9G=LgWs6IV+HtaybQzbOx@6cm<9$c`)P$F2xSpWrvm
z^lu)iz_K)C1I-(fxIX|)*g}}^$K|JQ30ikFcGbUW#W6PMne_1Gf6#{d{8Ck1ixd&n
z%(0O4|L_z0O<%<$SbUWZYme{A=XWiU;E5SFkifNP7lEcC)r><%-
z1GgMa=f2uJKkZy%7#rgU>R#z5xb);V6_cSX;w|E*t$3q0Gkx=kL14r0;0@Y%amQ;r
zKHR}GliY!yguby{g3?l!@TDh%`oW$+ZN`|Q3TvZSu`o1z>%)`PEZ_khyc(?k
zSlGmz)6Xx5L$wP)*%L;((U8!RhlUrFM
zr*ZEb8YS#s@8i@A0NH5yNx?1M+L(eK(@^joYCNs~fNnuEY`Sln1q9*`s@l@i_J?`-
zvXEZa*H3+GD5L8>>~INc!$30tcuMNgfWuLzkI}2&FH#qUe<#}`(kAs|GGNsFZ2nRO
z<|4k>N?ilB+W82IYOPUXeW2W$1(2gI(#K{sx^pTIDSN93D)C6qWHPBB(b0I*%y5H;
zna~_aN}0E1n?5JIB~)B(n;L-$1pXVb7D#P>zvTeY;x#0{R}sLhCe;e|0iF-vb;|Dd
zsGVt*?0V83XMT57Syq$e-9A0(MA^{6Zylh|XNrpoJ^?bGm=@{%23mNR&>s)>i!6=i
zYc}Eo&idHZl4o3_;xzkAb0c+*q=%oUb&y9`8`~5RA%Eb>X(#Ew-dU_2bycRW+`arf
zry#LhjM1^7pn$*sVV!p!HKo3rLRjsmYe;|6E&CX7#D26+tT`3mbpM)lgH=dz|MfR1
zJq*VE=(Ry<7J6RZ5G4alalQN|-UGoPs
z_oGxt?kTHr+redNzLf=37OR>R_n?DR_|>KeF8I1~?LLWQVF_tF
zPi&aOBGVvuTBG$Tk9?$0Uu2oIlbM`O>f`NhVJxciPC*s1dWHKOBtAlnm~0Btm=yM4
z9QOs~3RctImZ`CB5@gKFU{1Bu29&(NU9rCH-aM}P_=Vve(=(qkl3w`4c+6>NoDbp@6+0zWDTq9%P
zwMzxJG7ot7`4MY_IpoD|39_Sau*$wXw~`T}-h?eVvzvAU7--T`>2FPtk`%-J})T!z;sx-QVK8
zSFdCcevDUcR}3Q$4x{nfIX-X$g$sXFVxThlf%tR$e)uA%y1%Ztd?CGZckb1Z0`wgK
z42!nu>#TR*LLVyj=}Gw(=-{`&#>whWhkjq4=KX>TKV@d?*aHrd4~_ut&lk3
zsD!z*2IL-KHuvhik7I}XW~1okb#JS%?VBQGMN_#SEZ`4zgP^m8SSVc5Iaen$MHQkr
zBT6GH#1Ch5bO$a{_rdb=!b)167->8+ad&m_@Pel6{l})cIeTa_Hp40Z3Z4Ggr4Ttg
zr)qcTc*c_w>1fN9vu-*rX!Z7q0Ffh1K^NiuD0qOHawMkVWvBC8x{g{MtxjZscbGbe
zIAih_%inKg$?Pu$7p$s#{B7e*c>5d?5SGa?Hql3a5Zuv&Fs^QDRwZ|#qLr#bj}z|X
zuVJGFEgz_{nWAee{JEKp$Cvx(v5*j<+n}>jS$$-!V$32)_<;OI1q+mvc#2oJkNQA9
zJOLz54tfQNDIxCk>}n-K_Gku=gUh5TfojmmEgr~!KTO*^3;2JKK2B9Fbv;Zkqtmb(
zfh*zlJvBzQW*&)hX^$JfO5!cNZcLdM@~AyAZ)rK|o5d@0Sc1Dr>tO%(WM(zxlXACm
zo4(sXvO}&${uL@HKODeM4tL+1pQ&qK=rh}5XVygR7
znYPQ*my`|YW^|a70%F^VJUr<(x;|V!=$ROJZyg8gYfvRAk;nS}!J4kDZCPetx3Au0
zn8j$HK*fqmiZnew2jM>XOa8v8XEink7AohlPLSlBn0&Iwttk^U`V*0s?n0W+T(5sF
zC$an*?>FHO{ypSI{LHE2WlK{!%(BKnBA%)dly#ZTIic;kjr1(J)!G`tRI3j%K5AA-
z{r*ho+NH9y^}C>4=s?^WyQgrO7+I3bIMml)tQ(fp?3UY-{`xL<&w>F76O?QTc8Td^
zS>%3*qEsuB)snRYU|Iy-j_E>=J;$?NZBo@*XsJ^QUMw;DhEcw_$i(xg^bsAPVy%y&ay
zoc`%`w)Yarh$?9NgjZr=qsRB!C`eB-L-mkqI=mi%LwoH~FR5V5fL;nxAfI8A@(#e1
zW(Bg!UhEQ!!T$`7D}SWi|O6JN~V8s@e+qBu4RwPg=@PY*DU*^hB)%FON8CE!Uvj{;!Q04uH6e(>f6T=-Rt_RpHp
z_ZHk`u+YJPJc3VJW1j0R<(#O!%Fp391VC~_uac>}fJ%7gOO%+A8db?L8G!YJVxHdZ
zrEeOR2*>=&%BLn2Mc=>G;S}M(AvK`-qgOdn@?^x?BzHTksqQ%^@bRq~P!7CCm}!3!
z$KkZiSi4z*`%}?*t*?v9TGrD_5?G-$IooEnywUM5wK51-U%u#4IgXHgcglyL
z(SUg^Bl~7OjCL^8?+<33WY2IL6w6`F)-!X@&Mi^O|YO|+g
zHcE>TYAE^=V|x6qVmcwd+l*bsF|H=NFFRc)#2UuS+G-jXC$>(6PuG^X8&+S?ovGRR
znk^N|^zSpSF5xby5_dBk!}HMLghVGY`nJ_YcJB&qQ}!%GNm(8EF4Fo1yx@Jy4}zvD
zF2DO+*>A-g$QiU=RBp%C$ezL@?{zGFRPC6s_t7MN@`CV*bK#@;3%jEk^Ij~^J7Pa+
z8b`h~DYBz-nkALBSe1K;F`e`Ro>E8AvIwJQ(1=}_4qI_^qT`i!W}?Kd@g=Y|%q6Jr
z>Pw@Gr!C}f#T`+iZ>wvZNXOzy8aVpk>P8)JXy=~(43Yb5*iB)2IS37=c<#H|>9zin1n)iKGF-{?XdN=WR}T&iZK+H6KYw|8+iJW3-vD0kwbbveHr3K
zW{5CR6p-dpL||~DemfRu=^PU{Tu|&ByvmbLe(bPM>SnQEs+FxtYaB1M7{>Pqx#YiV
zd%Lio{_p(OIJ4joB@J%lb+CwhZrs57?Qgs>7l{a#xkAixNkf%}8CjLTavZ_i9%s|y
zXBQmw+G#n$)B#`VVz)D8HMV&%tO&7_)83Ze7yo?ukou6sd1AyQ8HMOiwpDMuBkbp|
z4UL@8ekp3cai7~)AITt+PUG9Eu1T9{WfinLo~aD5#xaSQ@Y#62ms4o-?6sY&a8$^r
z*a#N;!?wB*YF7efC=5hKYF^R34sr|X@8B@JiF;2~Qe4bej)#)i@mzZ0*u|Q-)fk;`
zC~)xZDoeFED8giHhmJGh2f^7??rC9vmaB@x^u3rn2Di~$REvbi^Q|4dGNXuqRYMpt
z<}`3b^S84CLR$1K%cl52Kzj}>zi;)%?mGK7GkruHkf+8uLr4}Mb0C1z?ACQ>Q6#{_
zCEj(VTRDrk^|v2ESI;do);r6I)0tWm#LxL6XmCY=?GxGjV0-2msKn^nWb}d4_K)Fw
z5`5+vDYNSGA$uKZa18cw+SR1pSeo`EOyACkd@tW~l8r5sv_>L*Sla!gk0B$HB#
z5WdVFIg9fzWrM!5P^4Evp6z}b&?`XpRe-@A8*_xNB!I4r^MT_xhW`Hkubf;>&QUCC
z6!Z`XxB=(WNDZwuF(JY1OF|Ia$l
zY8kW)TQ3&F=b{o45q8{Us4Qa`-#tY0B3mo4(Xxo5!eVb;6386K{UTKn$xtO1kL{MqM&f4WPW
zuZlLCm@3NgbhO3l)A+N45^oVd(*6*C9q$30MG%hK1{FR9qQ}Yr+!vCdZ~(EKy7_4C
zw25s(a^>J{bJZ7Ed$6O>*!SH)F0q`iUe@YP2h#idFtyjyOBlLo*+kgkR!u<1c*`$b
z`c4JJrC@+&H+)ix{2xgHbo-W}wAhPZr)#0VGhjWh@HJD7s%HKldXiqgB7HS)4W%h8pRjX)9gq9tyaQJKNEj507Vf2Y%l`Efb)Sm~R
z)R*aTMFBCRuFPjQ>EMU>ot8~io(P+JbwwE^&fU?=?e-)l4cTO_xjOjQ`YvA524n!d
ze@>%whY{k#k1jf`FkI<9*v+?qQ*V(^!Z5vq5OeDVAnsY@dtB*llY5^+29bS>0UE19
zjB&W8m$?D{bPMAN+=_p$4xsbr+|;lg3bPXt$#ZutX4c0$P!I2Ribk#C%CIBdTg
zAN@Ob)9Nl@ja%3wE+|jT4bjRK8Q4{(gompjHsqx$$x|?dwE~
z^o-FBDlEWi_v>vS)mey7ml_R21+z!_S+aPKvfE2Y1k=|LDbFY6B_A4WjO%2mrc`Y*
zH~!CIdpf_|XRUa_s8Jt)KsZeGmxQezV(u;Pxq8S6qsSgF?rK`}}+1l{+Lq
zK9Sr2Ei~@!l%xLkc{_jEt(3{#STq`j{wykYh_)_9i++rBT9ZrTX0`s@>zkr57c8uZ
zfRYOS^KkQ4JCEk&Pou>!r(WsIAmHyA>oa2xpwopGTi?tDBiq+?gA~kLaN8O^7&l
z(B?EURgZ*gM~ZmR@ofA^v-Cj;ed|t}w|oPq35Pm7gn?CY6ZAmgeZ}z;Y`k|Co#tAN
zh+d57uwCrw5c_@A#3t^TR4W3>H5W`QNIULqyj<=SL6VjCmg+Mn#YI-j%Lf#b@cd#2
zTC1r3!P+>;r)%Ix7`TJ-@twUzbOwjFZIZ{Jt&0OA`mkxY!1Cx*@%W!Nrlhu+8VNWX
z5JF5qHK4zU!81)IvKV&PUwW9n4Bl`fPi!F@>K!l7l%K<(*h^!AH1*D=-Q~q+_)TyQ+b$rJ@qaxcimc+`XzLO;-v$L}1
z@vd-c6QZ5sl03dQ;V%eU#?MU4Y&{GDWJ>qc2X&6(K*EtJ=|Gc>3DuqRI?<%QAqrkk
zGw6+p_nKJ|ai5_(7emf~!0DYIo(e^k09M9bRtv0%5U9PhiVMJfP*!dGk>~SvKzy~k
zpFVKhe8KhBCC{fIE<=n3?Nj_pn02(4n^vUHH(BrRI=qO!h6oOsrPglRNRRy7mzdyr
z<%?M9mdT2vsL)?j7-%r{JLsU6cx^Y!n8dW;WH3B;-H?Z@geI8L%%*lEFbJ>m_QO^@G*51;5!N3
zH+H&noo~#F8wWi@=y~^4l2-6%o0)|m0M3ViR!4h~3}G44YQX`i5c7F<)KqVK8x&fv
z2mX$!-pnq4+%N{(a4Y<%@7Y%B;?qm{?$K_6;zQqkp|=_e$h4bTh7
zEa9qhw@1EX1oT^gI)X6emc^S0VbLveh~vR<4vNPm^?hAhWhsPX*;LPkcjq$}ynY;5
z|Ae5g7ABO6*6A>xv8YkB8tr3JS4+`~UxzYBh0CCUnmhc?gKN2E$WiS7Sw3aoxHwr>
zP_JG}((i7FB;%pxTZs_%Sk18>+bqUDv$j^oc-p2gqR0Vxli<5Y5XjZeEyZ;U%()cV
zlP4-4?aL=#Q9ekFh8@Zlz3C}t528pe*)T;my#y2S{5d@#W1K-<#U$;YUq2;w?-2q+A5+00wJY
zy*-~#@eHB*OTB{c8)(PIWP?1f)(P}kpr!iPCZZ@Y7Y{*VcZUtk_7zw8qfd>KsszwHBONzUv8F0r~AdCcX|@ZI@$kFH2|wQ
zsGetnUXjn4)SzJBTp){cbSQ(sW9`WpN43gx0q)|ijRztuLy{&(0AoM~B`SGV{xYT>
z@YH1MCd0bwZbldJY)nu7di?bSs6pFN!{t-y2HJ75Dc%0GooZD>Ppo-Frmd_HDP@lREbUrP6-PMQA)H0wp<1!s8nm9EQ@Ho|(zxtTdr0lZ;h`X}Uc#d1^
zUo8}A`{3$mEyF?D%%*c
z!%?a5#~uJMn?zHj3Tut4lrg&ZixU>+`irgBlcW?olR?)%@54uq54X*kSe&kAw=Vup
zT_{j}^A`edr8g`uVQtttS+A1)Cz1Y-t=wwb#37@v+HZ<(-J-l
zR9rD#triO*JlLOWMY~2+{}K=S#@YP$joym=Vj$O
z9)gS7iXP|~4HL>%q}{BVCLq{Y9=Tz_^ay3HSX@q7Wz>{|db95>A@auEMf!{=@SjOl
zq2G&qQ;87%hP;LGl|reEXy6emyVRlh3?8qWe1~qg`A8AGy#og{jsrGCly74E&JQ*c
z0Q3qP5GI_C!|`@|WMQME&}-kzTb*wBwwXf4Nf4~+FKpQ8`XLk$vKfJ55$ffS?(u-!
z`aPira^E?qI)cmi^zvuyZGy_WIA=aP7`}i}BFObZKtN
zOO;RcX20$WJKtk7<(;Xy(7}5XY%}ORW>A$ho+w
zU#Wb3A3LK7-MVANw19VOkL|(?7trH3X?N>rwHEJWH(Y;iHZr+b6XoZw
zLefJHk%55Xh!m2?_&98wtZC8YMh(dC{|{#@2+(Vh5PvuVCUF`3iv~LVzkUK})%%^g
z6TzgnTS5w*PrlrOL+mIxKuR#Wg)p0h0a|HX%7Dt(v$^#hWK=?or;$8>fZ#DOebfNb
zy-g2%cz)knwqC!jt+TP`lNEAhR5~poOyLVhNA^{f98=6NTPZ8cPqR53zYm-_ej^Df
z`)s$D+)M3ZU`c6LSoT1-J%(9Zo3py3OLC2xN?(-QF0NZd=Dio)Nu!%%9Q1*y*hpNe
zx_v1PS!CqbWvM<+csj@v#oc%hM*2@4y99R{y>aT?sXY#)QG&MnfkBi#_O3qGqrIO3Q*TOq-mv(Z9Dp`c(9AQUE
z*5-VLKa6L9bVen4uf<()ksIg=9vqT)PxpvQo}hVa)JY~hrnYf0>BQgtqo9^aXw7=-
z2<^03Cd@QcZW`AGGp%?T#e*s{@DX@f-&ovK@J$TJ3#90*^GXREkTH{%6NL)WeBvc_
zW0~SO_4_=Bq03)TJ@qFsq$6Yl<^D=rPmIF@9UEg(<7mLXN%U~vo|xsEWH?;=>FY>S
z^T@ym_scVg7km}R-CFk>{_B1@q@LSBAaf^fF_5Kl-qPk@hh*Nf!+zvMeq4I|XTby-
zZn|WvHCLlI$+J1!yWhPBew&MZfiZ`Ep?93i^+eY;|LO40gw1t@UH0sR8vBBcMjpUy
z!k*-v^<}keww|#1#mBoCehWPxXob*D+2W0FM8Vyi84DJae1nChx>mN!qt_&THYa+7z4*j#ptpczm6$>c
z;ute62b=s*Nd>X5YGwjY?=$;UEfVb9a`t+Sh?7O;#z*wB6{0g7pAe+^*}>2ff}w9+
zhGM3w8qGD8noFd|AK&tS5ju!Fut#(*{zdc8M7N(`{bz
z6Ay>b=Hk{ak6pkm&>9~4&2%oo3oY0b=*q`s=ji6(+yBAunH*Aujq)P;s5ixIJ9_rn
zcCN7aVdgVBkn9EA9(la`f2Ss53M0HKpRi4?xl
zXvkBMr~gW_+qhRtbalz9=Mv!SRJ`$Mcl7T>ov=$sa`$o4G=x_J4+Iv81{VeA#lEoU1}w(Nj-4^w`&
zd3R+cd2%aoL%ROn;+I;M`I|)7byMk^1(Uih7T84q=wEaHt#g+s+vLxh$(5f+h;7=36jdC;-JMANFGT8?4p&hdDCdPZxIYj^B;nALi57)D-k6&EiI
zsG88pN2e4RhimS}@It-zC@%kh0R-vRG
zM>z(#IjoaFM9v;0KWIln)4ZSdxgMpu|0|=|xqkSCFO^ap^;(}`%&Pf
z6FYLoXcK0(rail}2NMQ6D$7g8{!1~D74E)zE6v#ISm70+?KOz;iw*0uEJw9NMSd2w
zt+(=s8qU1vv~mLkI+N+=liFG?(|wmCr?npfm|?U*%?8Si>Ax44)uVvOWd2{o5f4_Y
zINLT=x85ovBJSr@1isv53OEsAg0?=uGJvc?9G{hiM1zfoMqJk6YidkOj1OkL;mh}?
z9$JpDq$TIW%yA~@%#SbbCZC%w&uQ=(&&GRZXR`IWazv*srZ4tTTwgM
zg>*Ff?5_Gg)0TuK2`tZEiZnmZ)it|ev9UPu@G)%P#d}*lX{MIi?5mk>b`tHsD|gW)
zf4b@T)Y0K2dV{~B*-H*l5aCs^Lvm+SOzS~~2g`g}BNF%fE3n5#{6eqa5C9-W*8u-D
zl{~)zn@w!IXF4DNy7$d`tS|ni)h!~P)!XG;OR9yqM$r2Lz%IDnN>QyllnJ6f_>30>
z&bqjceRd(&o$$AY?}Cx|2zkm~ZYw@8@NSJ%I=jb>XV;&xOm0c?fz{2X5?q(~aBtiY
z1ifoFo(%wu^^F(&psID_CpfF)pUVN$#EB1LEpdgeiEBYX4e-tX*FRrJ+Z;6=E*6^kinMf{fK4ZQ0y{?~C5AH2A}u8_cfiNlRO1w}ix9(kq^?GDHYEDeg!
zKZa@7Y_6?YSTFEh!PA8f?NI#SsIu76cFjbZ=O5YhB<7V~6SPsFF
z-zXvF;^wp5-#0{)Ec^2h1^%qG3J}wyAUgNi7%g9{rB@A6;X1S&hmQ=-E~LU>Mea)@
ztnZwxyGA!vr4*`0W0U07CffeaQ1K8<;{%pb-pa=KK)tQ!IdYh>*k76ruRY(ak40-l
zNhvlBHgOu2mXl^G$26Q>ToxumyHh*FntyeCpveQ%Vdh=ESUKp
zlbzP&7=u$D)(x0kU6f#~N&DfY+6a?yV)UJB;!^_tIx_V7g|&8*iXv4IYtOOYhEUbq
z!`Md{=bd<>PF2Gv;+p4s-!%mFJ{^X;2N4N$-$dV8B5L6?9l^QUe}%6&(3w-1=1*7d
z%&3C7%d3;(znn<7(d%s*?{EKS;HHH2Z=R&TNeIAB`AZSc_y!TQlAA%EBo(9<+!brdqy^Mo|Mq7J*OlD$4mgmt?vF=yXE{V)(#bd8Q*$T
zQUenaPxmzeF0pa2to2ZX=?XNRP|yCa@tWDUy$%&2A0sLPX+V0}yvH1@
z!!G+Kl*M~&oiR$$d4aTMwR?7Mhp%Z8x9*HYV^ZMhvH64l-qoB^i44`_QQIIpe#O!E
zkpgq1zV@S@LDX)yLu_R|GF2q}{9SUUW~<@0XNyIGZEJG#jG^5(q+XdSLGNs;g-2}D
zSc{6}r&7U}6xYmJAI({OJ_q3afEIf8mA;NV{>YP}6h*|0&fxcLPAd^~TBBFe3B~bi
zI#3wZN31Lk`xTz$43hzPb(evw<~s4biXeW(XvBdbgHtop-nNwN^XV&`9m0
zRyrrlV8isls4SFAnHVRtH)lJOVr|gTW1?_JNvBK{Jqk~)_v
zZfsfQ$-l!R9;8ipL-_%zC8swN`-B`!@$<*I0Jx)hQpN2
zaq1DthVTrLX3?{<=EkfO72`v%MV`{U>7pq+ES96^k1scow)A0>ArzH?t375Cf=JwJ
zdnO=LcU3OPGEpJ1ly6H?OT`2rB8sOAs!_qR=7e7c_I`H$=-M_}7WlfSJ~y05B6OCa
zSys_Zh%JiMMX#3y0tIoP-c+>GZ=f$pRqVbil_WG!net*t*y3~Fr=YR|`HRw<+ur{T
z5`PnQk{Xdjm|-yMU*TerYrpcwmk5b_eNMn0o&m#a`4<_B2v0;_vBPV`;szr8Q#j?5
z#1b!x%p&v+Kz)Z#w-f|x7R6IVXnj4R*j8HEbcIckR)|^M#RSNJ#}H@c(_1RYeo#d4
zzzNgau2bvUF1@MhtJG)=j+Z;`Aqt=D^QY!~urm-^=!P5$?E@P*2v2&M6*FsSL}@))
zJEUF{pv$TXr(w2
z&s+K}B1b6Q4uM1zk`^2M4mRx47&o}E;}Vr#Q(I^UxCFTIt}lUHjgoQbGX%61L?)cT
z)=em?0Jen$d~ke;j#`Lum1}jc=t$ronAqV638r{AKnbtqXalp?Q)?e*Sekm9vZ4T4
zA~q7U5Eb&h;Cf0UO5;fkxGol
z7{s@r_?%C3CKsU|zmKI+xYDFkarJffM5)}`Xp5~ce6m*43lv!+l2VWVOrRJZLdU}B
zoS;O&&}_7O{(NH`J3tjOVa4*`v4ofr8S9(Twub$TSw^I-OqD~Sz0@OifW|X_`Vcx-
z@NH6RWPpEBFae!m(3k6%dXorv%ZJ9ujnf7W-XemITa)LC9$K{#Ns%9qtYVaBe%Wip
zs9WX@n8{xZeEWLyE$+2c6%bekR_>lTouQxYgJ~$@$t9S_YJV+45U+oa$z%K5KYY~n
zsX%aPF^$Ht_c9@&l`~8JUHvh0m?oSs>@^wm#&xW_AsaC%*q*+oWiKl%SX+3!#t=AV
zs?c$+--Of_o9N(Njd^&_mXCoDuM}+1rG#9lxU)UU-=C?Q`nY6M=o!QIF60>u;bHBVin03n%2xYhHFN`MFoMx)2%JAL++Mr
z)xv~ky;G5;ysVx^Sk3;&rDP<%-xt@iZhE-ByBHx=Qro*8=YxnBWr%2>g9T#$EQhoSW44>=vu
z4Hs!!P;xjOM$@hfpi=arSr0u6NvPyvllpU%iDzPW7-+f`NV^I+SoUtc5Wb#;E&@v^
zy*8i1T|r1Qgfi8bhBTK7pX>HA8>3aM5wyP|zA5^w}wbtmTH-jCYs8DI&WB3}?g03R;G^Z$SOe_96QGq-PdkhZ>h
Qd#xbK^6GNsGDZRa1(N}Z=l}o!

literal 30282
zcmeFYg9p&)Y2i+A&V@bh2AQ_Ht|D*?0*Qe@wjePvkQh)VfoyF-wm{heJOv(!fo$);xqks4lEiFn
z#cYAHMJy>vED0!AZ2xO)`(HiayZhGvU2JWWY?FX@i!D&6rNy=dDDRTQKuP~mNwNhc
z*@`6rl>jpL6@ga^@R!sAlq*R<&6Sp+9Tt!+z^SW6O6-q*XQ_ur+Z
zWu>L%?!Lv!N()f=4!B!sxw~6Q5?e{KU10J
zmE{%>6y?#|$6xuHT$#ajTrw~V=gnjOLt0cRNl&ZA?_fyp@~~K5;9V&R@YR2}nR{o8
zcjXU=_!dQ(z;(h*V9+QA`cnyY0i&nbYXA2m=t^ip{+lbUs~NuDHz_-vILBR~BKxp*
z|MUI%8!GDu23__Rgm9zu8qv&7O&jNK`m5K
zl-xrnibE_)WgpO&$`jUup|Y7GM72~~`8zprv;4`GpgKT{S%`*7Cw`l>CFQ5mTMt%=
z6Oniw<&BDGfRFMB4_1Dp676)rQ9i`(5yk?#nW;7%r>4xFO&cTPuKQK~iB9R!Gjo_r
zgN8D_)j6g`P6(t|XX^P2QrbrYvr=EnehqNtyqydMnuBF
z8V8mk18c?Er6OLIRQu_uI%q!kai9b*a`HNOx)ehj;5s)}-eJRAa0rA|IlgbFUIz9R
z5=!BA0v?k*?Fv63~xYOnv;(iEKoI(YG9PGLMrZA!-C-a
zUt;~X1LAj@{N>APM9w!(AchgMTG
znpjNuK&M&sI;J-m2@y>Y&4>tQ|xc_ra(&%%e=%&Zs<5QBH!Yz5JoJ
z2)FsT8p_!7v!3n&_2{1=pQM;DfMHoX-~`yhD7nHI7=${lZJ6zpQijxF*H?j?oHXSh
z-pJV9k{7gT8ovnx)o6-^KP1oDNfJS|(Y45-Uk}7swsq8N3{BC_xlw(wB*H;rtFeZb
z(NikK`+WYJc!c`}k9PpI#*dz^fT1f@iYZlzRRdG#a4Mhu^S1gJ$F>>6s4{E#$FqPw
zX?TOpCsX?4(A-r$&3FDyY*j$8Cfqu0lnPlJe65G(--ISClJ7Ie!o^t-dWj#lox5%}
z)`yzE=})l$H3x=2lwL`>A|Fil2#9@+r_p_)v$`l&#JbFaw&h-SdN^u21S$p=iLd~D
z{vG1tJI8b}I|uJW=^&d1Y+Nky&mVp
z2$Ln?;{YMCS0+!!Du-@lK5h-{y&R3Iql^;Kih}Mh9t*qx)ycq^M*n`e>mgMW*kex6
z+Sr&pLV5{!w&$dp=)*iEP0laAf&~}7nZQKr^H7&}1A9*v#RJ}v8EL?Ti{d6n(urK%
zz2H+Flz1F6bWxOHaF)yovRc3CNa>(!hoDM^jr6xx9knn$n`kQ6WWAwUg6zwT5O$#B
zAh6r#5oG$CPpEwfeNFgY1I@Rr^d}?wBw|Hz+uIpLs82Tb{h`?I55)m_$J~6;;Qw!D
z&rEt)uFcXp2(wnymMx2{c9i3t>U*{$h~OL&BbY26wEuWN743fc4rRJ=xe%
zJ0=g=&7=$g1q1a9cwR=#uH_rD2`87ZSF<=vBG7oh<;+wquZ8w!YKk?aT&{%%dhIKe
z2ajn}+a3MhN5vzD*H!IIzKH~1hCo6R`AXRPS7fdy6!ny?BNrItu6MhYnq$?W7i{1A
zXCO7lZ=5y)t7sp70H!6$S)(i@>+nxRO)!(>f>Zm8yn+CV3%3!cOgULXm0aN#pFf@-
z>%MOY7yhOJ#_}n+dNeefU?AIabru?E{Y0yax4&^|uDr#X2~5G0?~7BcN!z_Y=HD3e
z%HVf{n*yU*ge(m(u}(8dl)*dzdzsxWBK;5`CqS_(B$)-`)Yei@m=8bf$p9ub*tPTq
zCojo)#1r%x4Xi8+|CmisT^{fWScVpj!1dRPz4YMyjpY9q^aw@je3mE*6(S7CjKpUT4F8G`zSU1<04-2*R6C{
zV=lyOFaPTBpFrm+MEFV%r{mP0zw6NQhSm8PC?cZ8$H%Ko!{~#up
z)g4kN2^YoDlMTp{XMZlGk2mi2aUbMuAx?i{Wk#Su+&F-QG-z`tYxE7OQF61I|LYFm
z{m}gLq4Z{te)Nllci+)=W^Xj$IJ)MiH=9P=lSg4iwT1lG7@eQrFG<~~Dv%LumfyYl
z!6?=@d#?BZx%;3xxyL~U)}tI$avsrIq*7(cgQPET3{NLQ)n8pKUe`$<3s3(}%(FSn
z75Bw0THZJv9M9vNTrBueF4l-0%yz$mr(#e&i!EMJK8LVrWOKz>6Nbk*Mz8MskI{K0
zDDGPi^%@txG7$trn^v~E8QZ(;?wKT!do}ZGMjPG1cg}jFcBn~R&v$35yw4qI=iPZB
z>>=wRLEbp;^sO0BRc0d?5~3wPuosc})L7<-c$jF`RLDEymT-mb6Db1Qem)DqZAaCq
z<$Od9QUfe1#xKkt52J#Pg=e|Jl)B`E5Q%sB`N|D76A#v7Pp67UM;J72dcq~$m#ntyti`CsV@p%R-_6URr+JY0eYm)F5=4Aa
z7JEM+#F!DfB@7gHN+0U6Eve6`&7`-Z7rL{TRp!l!204G(Dk4v7w}-uC`mUYy|1EzH-Aq}pk=iTV
za1hKb$KlTx(F7AYLdDJPONA_V{wD>nTg6_}yne>E70YZqt9A5*?1J^_oR5`dy8$zEI>Sxye=&be|f4MV-4b2jZM
z)EKJhhXx*-4ThPN=N2MrY2nvUO2puMBNa(Q@QZfg!ZnN+zp(FhjZJj<%WMX5QdPsm
zE8CeAWYC_B?AI_Gz~K~h12EF4(HE0aWpWH*kd}z
zursy5yPeWPDijNP{zv70|AoTK!6^k2zz6nt@j8p5$u`37KV|KcX)m8|MhaG0w_ci
zkcs}*nV^qlb(qW%+P>6;fH~vMZ*wRhYlghj=6y@3fv;xl~>KeZ1cE0^VWiR9Wy8VBoG
z1hM!&bF0Ki>}sGt&y{owjD9uNwZB-OwDk`_SNQ$zLYc9)6+`;vEBU3QElL)!NwZdj
zA#dG^4HXL4>=r0yI_IzT@-+Ci*Tf$l1?DwWc8oOF`(a!0Sp1L=xgR6jwlT|Pu6S=E
zLmV#bD!`jENzc_GAZV|{h;BODju`zgvY|`Rvuf;h+f746m}7ImA4IA7#3-!md(A8DBwraY
zY8=rz5e#Cki!NTyRB8r~6;qqnWE`dF
zmnE~R%!V%N(k?{%Bz6jUbJc~*z{JDwA*7KV^9osU
z^@llK4!p6^3wim;H-~C{!%9@#KnDX|K{Xl4Zs`gjwI%Z2wW#dU|oNi
zp}3sI)fkF~7TZ6rdd7;>pmP{(dD*++io?sxG%+-=XZ;?oCI=Uw^!c0m_v}D*`_Ce=
z^}ZPH_DB=tz&Di9HH8v$?!>lNW$rfZ)$CWF
z33@+Nw|U!4SF*r)%VWH_Ma6+1{AdTuRDPfRs24$MbvxFF3KxoOAGqqgfPoI_Wu`ES
z{yF))(bxc)GuZvo23UOCrR92Sq@TYrPRy#rgT$P#V>_ItR
zLOFpuApSZLq84ia6FRGU7u5>jLZ&{&XMM)-E3Ha`z2^ExRZopzQrGRJ**Fq3!s30N5&p8h@jHj9%%x`>%m@|zP`~~&_~)BE)retJO8<*beXM1WgA~igmzmtu
z2!GbeZm|!nz9=~h+7O5j-j{@~JKq;~wA-;594^1?HAFeHWhuRPB)RyzH5=>47xswX
zc1UYBf2W(+Xnt%Gdkj%^f&QCxk@8Y-2|ZWrR)L`W_?iikCzrW#asA!+dW79>I8S^3C8kK58r>Sdd?QPlb^M{bt*1d(M%`q(
z(53nm#q1yeXC
zEi^GcH7FEoe{Owf)$%m0FX550XIWcBa&CTHODZxg#dl2=pFw+ZFa*eW;~i-c|0F)pt28-B|lJc
z4~?mP|Kl9bz$8->amN(UKj}03*@)w^A_GF9s^ySAiHIArkA2q^
zh9vh*C>`87ye7HyXan8Xf70T~4#G|B-8C2*aploBS#?{k|siUesR&B<;&h%udAZEosljdaAMGpjmY#>qn(5X
zp3@o+4RE1KqABEDiF}7PpPP4A5(NUN3^2!#>bGX+25AvHx7xeKT+-#TJm9Gy)u;te
z)?MiHCXb;uKG2Z|vW5}Ii3?!O@JUA+@z8JH_cEN@rrx83__Ty(>{g@mAZ_X>pL4^-
z5G(33ug-u!#?0i-c$pOh@4F&7bg(jTkgGzkJwKO?hb3Tzz7!?5{nC=G@~kNV-9c+7~s*A|dtf`ru8hRAVpzGx$dV62l@oIpQrk
zZL{th2X?#&HG$MWqY9bsgDqm6ePlu+$Q!N3$k}ePYYe3GGhior19ZvVOoda}(j%0#
z`NKSnAzoulC`$btd-eC0Zsou^$9+c7@8s??lBS{#&WvCGMQIm>nJ4p4S~?$n8j!pm!UN53ESJPX!G$842(
z4~-wdJKvIde>PNvDH`>rx~me~*w;bKDlKjXD|n9v>!|HnwS-gqq&i%Zd6x7IKQcJ+
z10d&(DTTW|`c}GG^RI{@Om;e#*LPnDcD^&tN?hx>#Cxmyl4dU)a=$IAkx|Osc`x%>
zPsKfMco++np5Gi(%lrNXzY^D6;yO8vc;0_Pu#|-FB2sl;{ox8e&?0S5rd~?8Fm5{r
zN_Iw{&ya+A%wE}B7tu=VQ>@0Marr&7hT8J=gpVtQz@H;?~Qjj)64-2c*!jU2S$OkLA
zUG+$t0u=>VQ=FK2`*t*CGmSAJ;MjN_1NBjh{e=&KR`m*`GPqk3e^{`MHGl?_UUYC8
zJI;*kE6*WTjFKU5@~RF1@W8}^I7~IHaiO65{pW(51!|Toar`xNfA+1%Kqbxh)BM#x
z4frm`2e0R28Ov_zu#rNikl(aZvfu83WptT8xh{922anB%Oa`BHbvhGS7&Wb=-2@p*
zDHvjc3qYqQTAuA0n&DLuDhFV>Rpf*TAK2iGS>G3@2~Z}j!zESfZY`eq)qjoMH}&*k
zV++5QWYOM=pKaaiNxQ_o0UcdBRY_AEq$<=mu&KDQp}Ic4NMk5G>&B)w_hf$Fq=xrT
z7288tH2|QWyEp3UMUC@?fy1Abupq&0;RvdN3am_gQ&QMHf0XuUthq
zS4&`P_%yaS2w2MR=WOX`PFkAOSlL|R)>*kkYTvV~tL+pIzj=G8r^t*?Y=ir?j$MK&gsbMDb(kgBF-3_ISUF>ND-d56kixkU7g$i
zU6z3RJv$(dU$#re6?FB%>h9UA;t+B)I4`0y9*;-k@WUP7V1036Z_pI_2@)IsesL>A
zA8LI}PD#b~B`vQ7>tQ!f&PhYC(O_LsL(4I`{Tn2m
zy{QXxZe|aPBGvav+N>Yri&Im%iM_;u*T3bz;KQr-MY5&OoW#;~?~ZZCW36l&e4UBc
zQrK8)MN%fef0p{#GySn|Kd+IYWZ(kSfayV%e@*a&y|#jR{f%
zWRhxTJ+-uu1l21274K*e-q|XQs+ihOHXdU3l#);-{01E$MMXuXB~Ex;A10%#h)`OG
zQCD|Q2O{RQ#RzLbJt_BVeVkA{pz&YTv^70_>kI03xsyFwZCaH$gP@dQ!t9;&;b84I
z_!k&cfl@hUQX-^Y{wy&}LtTWPm0w7Y>#Ur9l6Ts?i$0a+U3|_2))sJo;gFI4Tqr?|
zR;8T{$(9$YvJ{<6JF#niA`mzHYT?P4eRB+P1k6GaU^wU6ZTIU9OObEnp(;CKj;&ru
z_!v}Yih#o9xNL(#mLMr{gdi`#j-V!+k2zuOPeV7qK7}hSu-B2VmwzH~sqasZFvW<{
zh+}^1XN_zYX)Z!#or}seH
z%_e(HfiE%-_R;+^csFvckA}u#lDKpSV+XDzhS0d
z{jKRZ0%^;|+FxqIUWQBrQB<*jKP-I-$O1?7Z_2?FuQ(rmDHJR}BPKN_{N#QSd=g13
zm4UGn^j9k32AAZ2=q9OG3@E&OMp=)veq(Cd#(~`f2IB_ZDV0M!cC00gCSn{!(J@?h
zho2g3$}0Dmc;9nLc@}iO+lL_Z0~W7xQ^U3oSJa)Q-r{H{=7veF#$s&>It(!WHi)tH6oKM+8{2w8m5ZE`v66Or=S6<
zgEuaxIlf#fisSwtOOS#4;|JJ&C*`w~+c%+*R&v09x*3Aw>U`w*_^WU@6I*C9{km9k
zmE`+uqFVpw>F_cAZelz~)tIHBa=P8K0hM#5-*3tUq99w`Yd#{v6Ma1Esq!o6k#C|Z
zx-nBfasrQ;`X;S9cl$pG1JnWU>^~JK#Qgf$!D04LW7NVa<@Vs0%gySKT!_7^pAF~5
zZbdka%8Cdy-V)p+#dGvo%S){&527l{np?Pwm`PJD+X&`1KkI(u;g@6OuabTvz+{)~eS2
zbRVvEz{95kcSKD}Z@%;NNGMqlk`a;{T={vp^3n&K&z>w3-P)
z<%tolAxIRWTQpZ;gx~6Nx`MFtqx4D71+$UWwq?YwD~vXks}X$vh(b7ASzI$k`E&vZ
zSF1|pUAcDXuHo0}PE*M&p}veKmCAb~+9AX)r=AU8oDzN(wKu~Eut;m?sw5-Cr@Ouj
zK04@!?-7hLBdYILh9F56Vw1mQizo(K^H@F3;8BTDGhqdB6!f>28+!Yil*fbvv3LG*
zY&>S+h7=GpAbgJ5$BgRRwG{MkJrDSzJXD-!nF(vr_*-{?iS`o;JZ^%Nnu8?|SA58j
z&h&uOoLU-*$3fuuF}nwcT@&Mr@voDV5^
zD+he#H4spejZ-;DSznraoTrJ)WWzj$?Y%nMoHemo%>s`-vMYv*OYmIMmH*|=p+a{U
zJ9PDnutv4si?zk&(Fp0zAR<*X(bw@rGYMexHru?Hcij)#{
z{WxWrcbQTe27`16<+Frg=kA
z7?u2+;V)LN3M(L#c0$A7OHhBjo?vabw+&-_J)aGNmbv49-g?M0!+~tc6B0n
zbougsI`_E{QaU%U3gK#eS~YXY(y+_+l$HfinC|G*OUpuzYx`b&z
zSqUacQC<5-$HFUc1si$FCwu-wiD+aA7K1FI=Qz7nP&?=l|NHX&M{yurenZmYgI?1F
zRH)$0o@jZVcUpFSY%HOb(Sd?uiP)a-V9t&$VZ+zj^i-#ZAHEDl16~QrA$-2|v*^f%
z8wMTo?aCM^!AN5pq}(^rWglM$Ddam5b?ib)#HaF$0+SE{e^{gzyD+g$%jreYfFwVc
zY=ow{yW0%7pUz;Sm3WgEp}1kX6nub9MpEnY_9wl>Osy(d;5lL`UBVBmd9!yg7}A;<
zUgD_#DH`NsXGlaCGTE5U=!ACx8`96^j4?z+5SYqpK&DjLLx(}MdB#-g7zYCPd+
zZ|v2?K)dkMlA*#;Q$@ds>g)dP#NRSc&iv~yM-8$Y$W&9+FtwKP6m@cW
zqm~yUgj@4g6Shgxl#lg}KIVzzOC~}mY4Htrc$G5&15~}8;3#<6k
zX?qSo>se9*}mE76#-#
zz#in7n%o6@zV3WsK=V6P9|J8+MIN*th=Ss_<+?Hi@RFeZl{i%7h|Pdaf5*J6$RLcZ
zJzxDfUck@7U!>W@wqxzr(;)*}XUj}0A6Q^p-}zplhwhKr2$OD_eqIw2*8(zn
zUXcSG^nT0*Zm^L3=9>d~Zeb7tK|Sg?;&|4tdi7hgf}l-49D`v+LjeYbcLr2{JW^Mt
zqBIMIh=|y0DZyT9GC@x|UT?UM{ME`UWb4;&OICYD9h+F*r6nd29A;9d$%GL96j!?W
zXXcedvDx#|#)ru7;4qwVm#x{07v2F4THOQ1R)lpjoj5`|)~xUF-5PGvIrcm*++R7#
zM$O@0`4)7E2FrAOchRJ>(e(~+fOc^u3Wvn1IZAtfWi0yD-`=p*uau`)x5sO2W!gp5
z(d=~oT^PG}yS!vfZdqA60o-?(qv1UK_4ANBh9NTs5GnTwuTGM?olZ}KYl+_EY#o^Y
z{5Sj0!&3%#Is*SmpTG`(>qm)eDyeL~IKzt<(lbI~@x=>#V6#%w*}os6tz+@N;B78M
zp^(tJEA>f}rLz$!K%Ozs2y#~Os@&LO*V{kJHR{~NKZf6H{h7g6YKFL$`$SJoFt2Cf
z#c8($P~8sS=1VM-wrcsXZdAxZv)PJECmNGCotAQ88
zP~CVQ+se!`$nBu5_?__?FJ~@8b@sP8K<8t_{S6*vW!L}dx(iyj^@b}glf98~Y!It{
zjlK=_U8r2%ZFshr$&5^%5PGSX37U?TIA
z$nnK)w$p@vfpmh;WSj=MBxdNy`98Q5wPL+ld9xs|#TO9_0d~~;U$x7u3J^6&TS=}l
zdej3KkJ+8UR=aRownC@$pIHegufvRbSp=g*fDEmBmPbs^-r3T;n4GiiC(J5sP(*rV
z@ID83bnxg4^zi*OG;I0Mo`N?pyrAE#g0BiQ$woKo#4
zO-*T&oWl3;TKOr^RkaRrk4f}^>C?7&peh;zu7(Sl&ODpL3XENkB
zO3ZE@l
z7QO~@i)7Q^b>75wOa7^mIO?E9gDJQPz?v{Z)lvHAc*Ic)cW~?Y7pu_J-TyTB#LMSQ6$+i_;e+aH-(P(b^zAo`A>q_wx-
zWT3V~1GKj@hbwqB4Inx}lBjb6;u!~9{`6sIH$8kwjb+rA59rZu*p)~?&bHOxjfG+I
z$BvnpI>eU##HaMHMCqZ1GR%>#mGNmJoCqBSc;y-eB~eT)ly6qp+*IEjL_qTx?jIS&
z5n8^=BiHq@)%-G6zeW+R#wiXN@Z1>92Oyej?V;T*$$!QVRNHd!+eUWOyY(slf!AyS
zIUL|+7DP3IU|`H3+KzX@Z4Ud_=V)M7qmmBHPYb>pcAsS?nvyiruprYv`WGLCdbAwS
z^-hk9fi0Mh4m`yqgkGxGHn|b1<%!NaQ;>+=c{{AA)hxmA$xxDK7lGr|)+DzQ8(^*3j-mwLb$#D|?~eI!pgQigjv4dmTU=2lZ|w5BQK-gr
z{t-d6$5*dvweXuigwc8gl^pnQWOL=`=DP0Ef{Njjl&)6i94SluzNJcsl9vj}65z|1
z2ZF}xu+6{Y*q{HKh(~aLXxy-mqlkH-ir43;SObR1iW2|9O}iucPp;>KovXL~Gq-Zb
zw)JXcO+&>+zUNYHeKtd9Oy_35;smXJKPGj#LfCO?-rer|ESn+J|NNM0G?Id#C~q{q
zzQd{_m=Ew!Qi_24@H<0x96f36)$b%?=GaMAJ{}XzYYZaD#)12nP8sojY|g$A`}uME
z+wZRfW6wH<_y%N})gokM#g^vU>Bi1jADbTO2s%Ja$m%uH98~n(RD}3j8kKn8^hck6
zPsF@fB#riEOe1mqEKura4%pu=Up4W=Rive$28%2WwfpSJ7*#Kmx162l@AP)HEYz8O=AXg$JksG5ol;MXhG+QA~#8+we
zXaMOY%&ba=2rGUGq-yw;;|bpIvVd!!CYyYd0n>^)1Bu2)_=0`K4MR@x9pWY+0$~C;
zUI1=nF){K2Z=@ao)Ua-haM=o-eB$K$Z
zB>EKO?}0yQ6uPHTn-RbBEc-g((*M$S(;m&PUsM0ov{pg>aME}W5Kkt+j_e<6=D?N1_Y)ml`SmZY
zeCu1Fdwn^*3Hvc*r{pEOokNzpFU(=-wUdSd2lyx(uOgK^YC)>NgX4;5BhPkn`o`Iu><*TMbQ*e3+T&hw1`RRVx&m^@f|o{E9*kEAb<
znGt!e<%yrG=*8SgSP=N{?{59{XVs#ZVw|VsG=r~Hbel~XT%ke>j|TSUwylwQ;I5B5
zyEjh$1|bJn1+*7?m1$(PkI7F8VETD7q)4-e#i0ie
z0f7V@SG14_Y$U!%_>2*Lv`*Jd!CdRf&9JO0dUS-}7{@6dIgy-RcMm2kOyIN>XKayv
zBNjU}JMyg{Ao1k}j+`BBX8UWR8Jq#wU*hjLjqjeGr4zd*t?$*E9?vw)){I;Y^>zqs
z{cuOd*RC)r1{x5#f^(%Xmw{CG!-orM0?V(yq(N>9A2(IIPDYm0#K~*hhmhHRe;o@2
z2MdxI$--tX?t)(9kI9Y!jzZf^({5>vp{Kk?6$-@=pXcb&(4QK*QDHSHuVVL8q?HW%
z^oaR@6ct56t|*a8z67j8AYZvEblq2aSj=SBHj~8DlZy>(V_;GaolHdw(>Or|b1TVZ
z@y%Yt!0Gfi)D1vxgP+&~8gDG1+2pGh_i7oK?SxDBn#fYqH{Y>n12%6j#Y+@iu=sZ;
z+{z?JK;2PjI8_}FR{)fu8N`5ylzwwKS+MA7acXQpgsN6KA2^Dn2fUSBspFr3Nt%1{
zLXYGGY4Tlv3%FOUE*_T+x?304%zSxj*|9Z8G}u9YOoRaTY^j
z7M*D(jw(9avgK*~;Cl}TIr9Ncv^1&4njLUT4%h-yszuF$s_?j%rI4q;mgZr_)&)T3
z7bxod(6IL2bFA(kCQi!d49VG7Gi{FM2LaxVF7bIz@Q<6V!vVEy;?k>kw(Ewk2ubNd
z?|}f}YP~)>LCl8@EcS^FQ$v}sZ;;qh(F0_R5BK*0tWYm`B;HdVZXR%!L(4sa_pNF}
zm>#MHq{VG;{RXvp0Fbtb2g!EoG%%${n{rJh9+LJ+cWfA`+RYW^;}^1R^SK;=S;{7G
zE-Y;28lMEeIEgOyKEYVJDE7e)20!E!Ib2+BL1oI22{FMR^jtsv?EUlLtYIl4JX6J1
zln$~>C$Q%Ny1o9+D6a0~pX{>I0e?2i*0XPS`zURmLBh}jy*CU8(U@^H^wjpvsbZli
zukEJ*83wYUDWiPe{&9%>B#9I@X|q8!lJmLlXso!bE?aT?yY{Oss&zUT$@w|CPngN-
z#{5_z*!FQ7upuy**y!*o>DfwLwPV!XIN2l$*qy^s%lHB$PjiJpL>^Sq{dw4;@ap`w
zNgGx)ECl;eNDEdP7<>IaTTKO>QW(u;H&xL&jOM8xF!VG#YM*^hV6h_J8p)Zf7f$An$xvN}n>zlq^Ag
z42hc-=)tEt(0Ug@p0)C51QmcjBRF|c0CyiQjiYw}STQ04
zqh)^glJqrE!=~D(sa|au#C|^1h+&)jW)yE=PjoIOw{)?lQ~x^+5h3qjJ@o6dBvRD%
z`S-IV5f+%+ufj1F5yM#vXgBL1UN#=Q^KqO#{&IgSj??j{s)y@Zlvr(4*Bqf=X4WrI
z?P}j)1-X!}tNMe((dK9$ce6*)r`y{X@f=v+w94dZ!#_JFvMXaM1w!q8Fw
z@wcBhCvTllPDh9JKi78f^9PESEug^=5e$)XuX3nk4rS~@t52L2dK+QjTx7_*31dgI
z`x7??hXi3*3r~}x>V0|6zNA|!H%0$la+*Nq)m^tT;lH~~cv^pwe`VLWnrBU^T_+AK
zhhQIy)?Ov84@Fk!%QJ$upu0syeo~~(YatfR-}tU6N60)z`=Bi8if0G`kKw?S44A9e
zo?$7ZtEDgRkkCoQ(d*}9cCMSo-*nY`BR$@jXqQhC|Kww%fneS)BMXbVK1qQ_H*9*%
zBz9mAhj^Udgb7?!!;}|y((&u-*tJ12S*@`+74H^pH_1xq8MLuXbf-kQ*Too<W26Qyd#;os^^4q&`8rvI@IU*XAQavHuON%x;caQ0?;h_x06aXFYK(
zIqF2Nz*r8J&TB3QL!4g#8UBnhnF1^YA+F&f0dhmIj(Om+gibtU@n`TE@<|5Fy7f;
zY>a#f=+W~nlzQFO=Kf}|Zn?+WYh!%NgM&Gioh$u0zd!1f?QW^Rzn%0R0**z@j+5S%
zfWe^nXpwyXzXyqtjCRJDYfIfmB26{fb9EV1s0cV0IM~F1lAKReVWWX&*4o4rrt^t~
zQN5
z#_*P*ATW1TwHGFj`bAin;lw)JIR>n&9usrhID;m4XX&5ga!r1xe2jvrSGs2H^AFIP
z10B(DFPrc$mTm465)}bC@CNH-OD>V!bjLzuv4r>Lm3HKlKN=^lth~y;pZzzKLX3|Q
z{R<15oVQodWLk|{MBXje8lxo6J{b{8iOLJ6*uMKciQIVjz?_k^ctstHLTSj77|2nk
zm~&|JKodB-4>FnJ+6vU&CS{s#s2UGlS}vkIe?Cuk1^Jer6)09KuoL#?xK`j{SF#Mh
z`-e96RT9hiCH-hok_i#dbk{Uquu*a&P^38Kl+N=lD*(SFC;q8nR(LL)COR3w)#-Wt(B95|Jb|c}Ze3$aEeazmkfLwp`u|;RpINSN*%zhi}Zzs)7V_h`jf~FK%90
z+;IpT@vE${`xFMr_zZP>5oU)?N_6>r$1^)&TnTXN$T0&)Td*ub{mE7T4uef0Fq8#BE8xh4lCPU
zZ^XnbRF{}!KDU|NKffZ>**24|QaOrSOvj0M01KiG{N5Sxi`F~Q<3T+XNOxtBuA}?}
z@c|O6-X^7Md)jm#gvVK6%4nbLwCwA29K)9DH#e6rkbc;MBTg8%ITBImHKhwP
z2cLCy0r_lzMoB=cHJ`2Jh^I8Z?y?hUB)-{u%kxoDH#o+C?!bpyKenhN)UKnlR`lPl
zu6X`4t?`J4F}DITR&Me3jO3A5H$~YB0F#?rIr8oB2?3yrG8?2`!qfI_b
zsGvRFOZ6wRId-_h5=b-wVmB$t8iPcZ#FlR*A0@;rv`LvX(Ru%O$SB7G8@+o<@!$r(
zI1a+cx*QJi0pRdcej(C^$*wy;>l99SMj0gD?nn_iA=XQrtg-j(IRaAFtA;~c;PLVD
zCL#rSsyx<(l4h>71WqEzW{=0Yo_iUGXnJx~&DE*EzTG|eaw#l>jgFXc{)Dc?TC0p*
zkW1Q`Mf}={tQ9Cu1uR)I{*=OO!n=g@&P3S(>I&Eq@c1xv!EX
zGP7$Zw_n8kLim3ML^HD@yOof)lx@L7iu#02CPgY~;O&&aovr>PHq~khKl`
zWD3_0$+`05M26APZf)^V{4$f>PGfJ@7N1
z><_EKPoi+OG#dWAI2jjXaBvEqY*oG~txK!wIjjk6B(|b;p8vBJm~(;d
zIan6v?h>s(olGmdmi$Eq77jT7dr~?57t+H$6qnc<{^oWxk?{FDZ|ehDHUPF}F+Il?
zV+9PC^mk&8eYS;Et5-hyDo)~nvtFJuV38=3zo0Ys#HQUHEU5JMc
zwQ%CV?=|V?M^JmVkJFg^$hm3JN)B-9iWuaj@`8Nrm-BaDAhFAxdkskypskbLpEjwF
z)#nN&v&#K%4uiXxx>~uu=e6ykyaJ|!w*k{n&XkL)j
z3YCJQzyFF>PVVB>PY%C29OLJ|>_ojJ;^TPQx>l}AE4w+=y3
zb>#J6&^#u5#hV7u7?=D%cqo73DvON8thxF#K{$rdU
z?|V*uzpZCAT@VXmYax{4A*tnPZ$7vG<}W5;ulhI>a0~n|5qcmIloH>?xw0BYDXyeV!fV>1%$mQvJu^8Di@&+4y-O
z%)#LzKim!&=0l{jxH=5>q{gGj-0f_5pDMLy`{pBAG?Y!m8gGCrav(-lBCi5Nml#zY
zBxx21BvvAZI*I3&6AG+WB3E>Gw?
z-)Y;{Gd~>=e~fF~;S+PV?Z)VK-1^TPJjIHMzWhHOP=FoPn)=-NDLec>?Y;LmoL%@X
zE)kLFUG(0|Bm_b9=rwvB(Itc+7%hV6L9_^hM2!+HxIxK>K?htm01aatEl3xEUZf;{ii=jnfdx&
z#Xo7GbbfOeF>MRXZ~>9(cDzfs=xq8#_Og3gMgU*D2-eXH#X|M{=`=Ba=}P>1XWQLo
zWAjqBU(%{MHp{5iOc|ZKO-^0{)2Rz(kc3NdWEg0q)_S&IP#eQp=iN*z-pgkEXI+L(
zckef_6xJ6S5Oy*&@fFNhFS*NEL;%+n3dm|L4@xe_`h;O|_i2<;a3k_g1kHQj=sOvx
zz`K0f5IpqyF>dlwjF-`8b)#yH@IM)r`y;(3d8}zfCF1|}QCTB7g~b#Zjioo4U8j_n
z$blYNZoIN8A++$I6yOCg9sMjMw{9lWsWK45em5h7}4VfZgN%on}yFWm8v{^nGFs7B|wl2U#f
zts>_$SL}BEJX>~4reJjO(WH%HufSZCxM(d86Y`*e8I>j_@w|wx<1Q*^$Ac4jsvH2j
z6l6pK0gEa+9^`Y5!1IxG=;g)Y!L4b@J!>}pL^5ZMqj1c46nV#aYwN^B3R0yz#e10U
z>0a7bBsryXGbnQOmfrw0pf7AKQM&NI
z|5bv3)OjXkgi^m|_G)B&9P;_T(7oyH@U|v{Zq!;=B`^Ie&LY;e@FZ$3o(7(@j?@~1
z(=MSTKKA&wuL>P8MX=fLe+mO&eg9Kfs%d18dajN;p}G0=9v`}^QWkx-5j#R3<$HVa
z`~1E5<2<@zEpK$uL9qZhLPvxrX)r_UX1$kEzNC=unC0Ha)~T1)dWZYu;%TaaesPx~
z!T+oRhGIq)hFP?`b%_?+K1~c5oW&Xn?R}yC5N;xR{d>`Qs$ut=jv=!j<}8Ext9krN
zR&V3DEF(9-=a^%WI=dkF!SC{#XbR^T_DG|W0_%`pyZ`lKDBrgiOw-f;624iuc36TY
zPPec`S(MJ=YS>!k9vN;Dq|ZLXUHK#k_inK4nugAQDPw>kzrje4BO;P~e-Bo-7EkLq
ze=))v_d9BNW_eU;GOAiBF%DggzxW@+X%u)Oe7J~j-Bf#E(D+AY+fY@LL8yJKtzj`~
zL+-l#ukYM4A{A}`*Khq({BWCoNcy9x&!dMtdy9Em&R0y?%d?ALQ#t7-eVpbn1%7#v
zrT;feVe+m;UN(HR)nnu9Hk?goae|xPA4nw;g(}&0$@Xbh{+1GNCz>T{?|NCG)aQ7N
zDE4}KI5g9rUK;)#jeiQoU__3|F&7lDH%nKnB!9xHbiJ6As7s&B{B(h5q@MG!Lg2wtu`3KD7C->B!>_`Bq6N%By(LdtcCOFo8d$M%zL
zP(YN?d^MB^T;V%l^|mLtl}f4NBuNq{M&LPnzB7f`;8-lC*W4B83xnZq>v2M3DVN2H
z_`t4C26i^qR5KC>HD#HphJmL;N8_0FEa@Hfw^ohK&zXCFjx$56Yl>H)E@0Yx$a6cI
ztR(ukI;74!NV=z4bgi}v2IbAK&7GD`>vbm2tFABD%iddJ-f4;d`#uq|F`Ah$^%kW*
z!=rc4?;VqlYLO!fWV$(l77P6qXM#JOF%;haoqLm*4n6FXgY$jhHP0cBWVlL28TM~L
zknNLOGCD!WCL#*!OiO#8R2Hs6AE68OlInJsegwJj8r}rph?3fcQ{jReu*2%JB}Ad^
z{rUn60f$nfHzjY~1^%K1o0M2FY=G~K8Ez|G#0o9pUT$cbHp4nr8DHod<)i{ET;O%G
zqYk;IeEeT5-0GpQ{>IgHEE(S$bPtukgAU!G8ol+5HiNlZSHNOE6f1Q=pMj~R(L*R+90-I$0H$kr@lq@
z=^QaL`~?+|W8fU)L9Vq4$PHVf`*T}4H08gXOf%S2Jb1qjT*>D}dhpz`#k~QH-vCH2
zlLZ1fN#vZsDu8oL-6ob(EsE9vj@n7(DiH<_*y=T0TIbr!@KFtq$?56_lKh{Pf|KEE
zY?RrtSN1D-_p2{WK;kD8lM(zB5PO&RU2D&_K$2)BHxdhhTX5q5359!JXjlE%r8ela
zscDgFP_or)E}lFweJorxUKgk8f$dD8yra_Xw40S7A@1eL-E?(4Q9M!kC#{ZfuGd#Z
zGDvA*bh19vDz5lB2%Q6B_iH;Kf)FPl5UTrn(ZMqY`>9U7wHd^WsHl*>mpuQu-)u`!
zh}zX+^v2aEZ!rgX)Y|Gp;j*BR9Uqq2wnY7_WbHIvnYr_YG?--
zT!MxaNW=N$6Qyb4e^H5#hWYX9qHU~yt%N#pbvp)$lsgbS`m7B^ois``()J=exz64F
zYm)H^qqn|G48<$ndDm5z4%@<-8yZV8M*BEhZ_p>Mvhh8LFlZn+MRBZCcWk;?4cw^G
zxMQHJNYB-;S4B7gp6>A&vw36u5;S|#@tLld22^Y6jplKTuK8G$KgE_d<(GFtO*?{-
zxd;dTC+`-`2dd8WAWGNWR+ZHUT(%38S3;QsM5T;>9wz+qKXc>#wFA$e6OTWev&*eg
zj2g|$^cr8rPFIHzh|M-nCttpQ<{k!=NUGvu&vUQEd7x*9QhQ7FFNDMI@+d4Dj#eL2
zHp)>oyh!@|b(W3rP>#$6U56X})vRGZ*|8QTxNcVUYo038S<%%ss`n~57U0OWSsSo+V(u)pxB@cWq0PoK)q(Ejc?5%*V*2*
zY&>1|_{Q;-*r2cT*D&vn#3d|Cy_q_?D>hThq?vz-5k=5sLWqWQAW!Vwg-=fpj&cI^r+fa9muLUPpF4I9Ja)#YJ2oY>Prx4?C@
zujgia+0FH~OYtnP7yZ27^8s^alBGQ=5ghThps94}8efLT+fRQvCt(x$OHi85Ga--d
zKUA2#56c(w)AJsn0hm3Nbnh9IJS}fmYlZE|250YtPxjH`S)H59QQBgRol^`I`M%qd
zYTvBR(&pTK_Yq?Z#I&pe@3?-9lcKJfF}YW`7=){1mj2Gr{e{&)vqnI#-%o($U+NGF
zdHGL;M;cu9c0^1vBEdb{*#wX
zhnbq!t?nvdj|-5l7Fh;+X2KJknAx*=@`6jm7F>ZP+ua71I?eoZ$uf*V71xIdlR-&C
z)}IajUaulT+7PM>CD7j$t5hVuUZ;y)@p%k(M^ng^kLiz6rQZ7v8LeD7o3<(ZUN+}o
z30RM%Ba46XCU1XNs9UQFJFGf*H+LcVa%7n{KFY^gyUQ9Nve#MLHiUywOQHNs&{Us#
zI80|i3x^-Mc6gqI&2RdReeP%_rpS(UG&SFUF!rv0RD{UAs`^%L@)zO(Fq3k1H5$tZoJNn83jYOg_2}JHnutr&k?qg@yTUME!@&W0i
z&}wYJL7tWcx)sHL*WbdlKj)xXptgUIBbwn|mG0q5kZ+~qpQqb7kcpL0w=@|KwUJJN
zPf~lQAL*SH4p)ysQ`;PZgz`^V1cIi-=s8z=PZHkr
zYxY8!r(LZ(QddWvjZ@pxr&d3fULQLvY&RrG#eH(#eIs+#axu5Yc)D|^3Kx{gn{J9r
z#KVb_%f~UoXFvY-BFSGOdA=thzV{VC-SmqOmCW5ovr9=wWH9~iO?qx09i|>x^|HRB
z8e1&=-43NdD7(C)T%*y+n;}FQ7m3&OEzrtk|FwP}cDc7?1c~;Sv`o1_90H#r*CLyL
zX%;S_>u7>QZASqa{)FA2I7i1nGM$MPCC7*Sh;O%(j7m6XIqy2Q6Ep3%zEr++mX)G#
zOC6Plb|j1Aqh8;-F(HyZai?3svp*7gqE&;L10kp>ox&p1g#aQCd+#MrTlo6|*UQSg^y}NRpD8=`x>MkFwncsr-ylZX61m(^eL@xoJ%Aa^I%Yg8L?8
zt5}HXyWsF4OCb@pd2IikvcXvP;&AlU1JrWG|a%vo5Bw!_1ZJ;
z-T!VVhYPmVu!~fRLoD}8Qv+7T1oE%lcc0jj&5*O<Tzia9|a!F1k7rc
zF#7FlHhqe?u_FQ2Gu>TpxRr=T{4z@3vrmX(6X)(=)o`&&CpQGi#aHMceu6+0Caoop
zKgwwxJ`lX)E57A+x0tAI)CwR)xq0XZ+gv8~oC#NV`00$K?%}duzq;O{C`R9%1~1as
z#CR3co2{54Uk9~XV`ZBS2z={QN~)N)B&MZx`GX^I!6{Y8)WVgo+HOEoq>!?=M*Hkq
zR6^O`CZj-~d^t*4N{-7^0&3K4da(XxtaoeSWQbNORIOS|n2Oq$!g)_)ER2>l=|La>
zAiw6>8>l(n3`xB8K$60L`qD1@O^q6Ai{eV=+yP>5`sI1~3Uw0o^&ZksPaQrKOsoJG
zJ(pOeTWPG|$gqeTeL=XP))pH<0X5Bv5f0Zwds0UZA$5OA_mgsT2I?~M*{mm*;SuHd(`q|Fp
zxOU!IOg$BPoq)&~K6LaHgBRr8T$GND>+w{Nq~AE$6Dsdx@*vIEP}2j8@Q=L1mDpQ;
z>_X#N#|dfVoprO-)_#!u4~AzhX7@vd2!|EDw?3?`hszVaxDPaCLK43)RoWCui_?uM
zE1LB3h%BtrdoaYS#=gQozDnVIWh)%P^R6Qzz&tLipId4hfZuR88;MbgQ5?iCF@R6L
z{n!;NarKE{y_hHW$Zhh`w|cm^1|zBZ1eE8aUT5c3akIXxVv>_0*nXZ?s++5Y9Xt7Y
z{M_T69uG35k{z>IG#mYO4qZgpp-@>88>bs3vU%7QaC7ftuhtEdLnU$6h>hj+{xGQP(aZP%3x(W{i=TFUP@T
zNVj0bYw?Ey`3Y2y1Q5
zbd+CHU^BxgcPziYeH`?aZPk>pGDZ=gH-wh04?ifoA3vMN$MW>%Y_dPMPv7FP`f`A@
zspfBf6lM0da*Aw!r~R1ebJsBopkzAO)-6hxZ9Ifv=11qOFd86=@fG8`+Hvd{rX}Xkwd7eJJP3d{ZcD6^IjQF9gV8-
zeyz5$BViBJueRe>oobY;7iJm!9w>ULpNe@kxtP4ycii
zY`Vv?QEFB6j%1b*MgHBcJ=Fku8P6uEd-MiRkdhsP?q$`LJc|g%!r^%LUY}SP?I0fL
zM)uJ*-&mMU|CZ;UI2Uw#IkomNFBcN$pE+s&jGvHIVmz1fLg1(pECT*(Se>)}xO*In
z({P4{i=n9c2l%TKMGcD!?`?0Yi5f__l855GOOTt8M=hP2?3GB7Im^&HZ>F;$Kz+eP
zFiUduSMVvuy0z-(8gv_d)s^zxG+;?{F!=qIRh-_?!J2}t!Jdn>;Cq!I~7Rnmi>Q70(&Ep=qa$@lk{&G8J!V5Zyx*&MZ(rs(0T
zG7!%*e6*AX5GU^Cz4&z`$Gj}aS1}uYlB^Hf?gn9@(l(yc?p$nSBDZr!u+3PZ)y@b<
zWfpv@xdpDttw*sTfgQK{fELEfKUyjY4SGN6qPNC^;~jjQozxr{+%hT86Pbzk*9?m2
zoiEaD5GGqLqbsc|lQj=@BFW!Bj01^=?W3$=YK61!Uf~(1G7WhQT>yvy^V&q5w|`4#
zz9(Cw*;raN=u(;}L12Fu$V{{*Vbaz{1+Q7di4k)IsDJrgi>4F}c3fzE@0sLc!$!tB
zF6+Y?pAcsBI+vtNXP$rQ)GIU%A5lql=Z851h9`T`Swl|L4sS&9*Z!@$OFOH?pFdKj
z@b`iDc30cax+zOGYCWz)_^l&nfjVR1(_+4DE!vgDK++NYR}|fUUMWQw>^vSFR4x)~
z#Ipq2s&^QFKs1^0pFZ2bVIDCI>U+uguZgDFU1%qmt`Z;(u@Vdgil{L>u>}O@c%$zj
zRO|5e$q|1gR3oZe7W!+&asoAeD-5lR1*6jNHLYO=Y`9O4`2|jf&Na+s;JlAFJkA1w
zsPuSe*Yq{B8L%G%yy2f)$FF|9$+BNqwM0{gXr|Zn@3Xp8^%xI8h#=?Rn9J^xjN7Oq
zX<0|q8h%*|3tm3?Mg_2bf-^@=9BD2W%*%b2Cx%ty6}S#yF1CJ-h;=^IL#R5cYSRKK
z2TOdX2d2CKJ+WIOm;a}DYkB=cZU3Do1z|}DN~H&xhH(C|d1id9e~bxN7;|H(bvm;$
zF<-ej{?l)Vm*^=M{ys(?ijpXy!T&Mg=8#W?dDcV+XS2=47fYIycuZ71R1=vXUI%SfbVHM=~_qGJH$k
zj(qrOvv-Bwd5Jg87UG3Uq^f#}eg)Yt$9I%Zosu2LC=_^IWbE?q%)4bs)x^)=Rebo>
zK)0qht@C*1jLEg6Ac!|iEwY*X=|3f7@UW;W^;Q1rp~Ju+tGT%&V7WJ0@O}Ya@NIOr
zVe9qs$jet=0>t?SW{R+*`reX#BLAB;42q$zs0Xv2{Zv7<^yV-f`c&9<-}E_MM3f&y
z;Gw03>*!eMCT02--HFxw$v+F#?NYIVCxPARBN7Y;VLJE*Hheu@!RfDoN^tQe=~3yQ
z#3y-4+~UW)6Y9b_|Cxzq#TkseT<=-=a2Hn~yWJ61^u)4{zd0lHmuVormO9^KY*kvF
z>RleScJRP$U{?caVV-B({uAj$G!L%4Ui;N+e0`h&NZWCKL@{vN0C~IbUS2Kyk+cwR
z+QB=(8I$s4LY`7?pQn=TY{q>~vLM!hSF-bJykfEXZ4*
z85zMxU%JZ7k5b8#699-C?mCG32^#r+#H0T2AIPhrT{9`HA2)9R-$A@Stzb|Cj6L_G
zxT};Y_RoTk*4*Yjk5U>(=(r|p*MDb~n8HWC|HziBH;8`rTbp-pUDB5)yId-@%gVbd
zC@P%+w_NYFAj-fO+C>Lz=NNCnjzX49l997x_5s1$#80Toe3h1F%Em)`2_Nw%}fI6*<_d?1r`?7(YlM0*!*
zau>Ag`_-_`D2AxM-^1}G1Bp&nu?@-i%KKmI7WHFKq#B+w=9XRt=-$6QZec(m1IeJV
zi;%ustH??|&fccd+LJK3miWNB&?rv{tjqrttN;(&b)o|GLi|{8e9UYmJ3g)LI6I=V
zS=`08b}#qC@z`cHos)n!miV9EG=mulS@hoRnt?7`cZFwg^iAKC{RrzNIMjQE3(9yl
z7>xe1NmNc&<(jXa`%`>+oju5&*r?WP?yGf!iUTaw=L5=JwHkI7NHmgMZInUlKX`IU
z-J{!UKwvyKp9NCj-$>p@=?*qus$Gx!p@kY3O!}Va{`l;x+~1{j@WrvBX`=81C6wlY
ziG!fIc^1PEta?B0#52>CQnxLc2GnjzXjJvbBsL{q*E#f2*JNeCTaE--=}*`SOoI1RMJY
zRFstc#XF3mo2YH(N7!ZaYO=E^3zCtpHzKKW!lwDJ7TmQw`pgAJVk!WnTZ7tnuz!1l
z!wlPJO30*G^*rD`8yLJbpu0GUdHazPfE0iz6nEKYVhFKwQI)C?`s`wkm;kd}J
zmM)<%%9$H8K^R-Xy`EBjkAv+KG=1;Y`*o_^MUsbhr5F
zp~YhR`BB8tEdkLL;med_4Ueyl3>X^#*#aIzK9hjf9TLVA89`RdyseDTA-HphSjP=d
zNbiYAwO0%723i*W0tEiQk+30DQ?yG7$g19*p+QV}I3V-a@g&j;3mwPP53y%>PH8D+
zOUKE4(^Bf2mUf4EUd@_S3ixzxkZU~46CU~+6=85}x^ZCryLrZn9K9a>W<7A*cm&ra
zmF4Qs-~Sqpf$@V1EQ-xAi=W$Tk*j-u>%I^=xkpc$Nn%Zq1L(U~E@fm>nH>B@DBv3`
zQzhO@1oivpZ!Q+3xI!f5d*hog0Mz3J*namx?&RSOV7^v$zCmKJ&;=szy*VGHo2*=|
zha0g*AFJbYZF};f@$awb;_hoT^_CI^yXh3+<)iFl>*+{
ztvgzg6f;~Tu{H;tNTe`=H=N5JRPJ1Z7<7itnkoEQL)T2Z4_oZYQStKwRI8OtJ-8j9
zphgYjGXB>+p7b{qo5b5|UFzq89MI68l(wSF97_tiJqkFRmHr)w!g0!xfk=O^X8LY_
zw!d!YtsVll%pf9aq9+H9l~phib<+m=VSlu0VT
zYi<%7l7o82-G9wX^YRnnfl-ntBPy*aQ=)7hcF{tvUX8uo0aZSkxe|(i!u@&e#Q{Bl
zipHTcJ3vvNK9@ac$g(DstjGm5y0Bt;`qFx1qko6RGu9F}S@MZF|L~^~4Immrm0w}o
zzI!A++a1;vw_AJZK-K`eHXQT-<09Z@^5q>-x3PZUR1FIa%kHtYN45%LLpIa*Ag@GI
zgCt3mMN{jtxes&cD(v{m&h^_%5<6#l`2VjIF;TjS;Wvf!xG|1{zvtp>jpOv;)@KnD
zZ}*m#WI7_o@5rts%@~C8Q@wn+pz0?mB1SIM^Nv+hb(rIbrG(YjX@@!(f^1^=r*U*!
z5<xhjVm)766(d#X-9{Lfs53RjF)ahmibPlR=^z=SX?%rVzr`hY+O?;M~
zYM!;*c7BWrvSV0PB;htTQ1ZnclR?%Vd$LdrRc`*8>>ciEkpS`YcVwhHzSj)&
zeCLNV%jgK$-GPNsg9_AW+1;P>^NAl=px4y$5Tq1E1s#|%n4<34QB17tV{iJ{Dd|$}
zY)SkYB-mu}lW~N8ta0qH${^S1cqv~F7IsPY*~&RDr6}!`7cLzg_K06>>Zf6qu%-iy
zBA7FmRx;CI&WL^Dv5k6Ue{8N}m1{*&iI!kP^myM_L~Q6M@&z=UFre0~&larT#nAWI
z<3g2X#F9nx=G(A40zZRZto-DubkA*ZNg*FcpaTrsa2Nl_w;5{>xhNqn2t)&nkp
zIOK94(&Uiyd@9wa>6y6|Z`w!mtkt&AjzEu_5~Oo8Y?=9g7xrkhIbJ@u*38EuYt`}U
zj^di0x54B!A=U;G#9^ZgX=Z7->g!Xr+@YJ+S7ZaEC67`@f8We}mX1JuP!3;SG?JJ#
zhVs1-2t)F5n6nLQk$}Ded!VC8YY=sA7ML|4%TU_(u~T=H9W)#4c+C`U4tvfb{s0S5
z1Sn^YkLEPQ(z2-cF#i4Rze}nkjb&LgcXB<(j40Twguz#b!0e{MVmSt03nY!OOn}te
z|K{?mf`{+8xVt^hLY?7s1`viY!u%bY@OH>&(1VXb6ozZ`lOU;Bs2BXO93Vcs;wtmka8ij2Wtst9X+5Vi;
zCyjo3YMtv)vcW;pVbKM#Cj#-qzMuP9zw>!lrk99xz#&j81^z-YHqH`FH(YS-``6>`
zzzhL1a)Jr@iyrKoMmGV-Om!#S*xG6{F%2d3_o|G!^XHay$rDUJ&nWBU#%B)^G~%qj
zWAZ5S%;jt-eYh~20M7c-sBle*-nftbGM
zW|K~iD6vY)&4(5^Lr12E9^)#@Ku|Z+x{>2{zX1mT0o(*U!z;ehaa4%+4wM|llY
z`f-e2X;kCpcC5_!X{2}2sjND$VQtSPJZ0S&JX-}!7;4~=sxXJ<5{q~(!!O6>lC{}$
z?5nX4$ra8lyXcTui$_0TETLhFY?O(bjbCBFu`y%;FNvgma+L0sU-k5M>=ge@hw(ue
z{<*yP1um|jLK}@Jf^H+77RbH<TpYFIg~v`w&qc7D+6vG=s)Yl
z@Ud$NKoDF9o_Wr9FLRIN2ZWbmHDrp!4c#Mg;`+qS+gS1l0!{dpr3fGLZ~g?WiF8Ms
zxZn7=b07z@2YbKIHV1i4UeRB(wil7Y--GS@lYljAjnl4mBCC^r`
zr3e?aT3~0ly&a0qdJm^}Uwq=WO^zk$Ek*aEL0lTA
zGd5*O$#F^nJ)F9}-^N7P6*&Uo;)m|tdb^{k3y^xNzE1bHW7DE0nrc7wZ9|
zA)ic6wd5Xmrw}wbX~4C^Q9g~Prr+tI4%&`srV}y_rtnv6T?XR0l_)){X>@q9(B|f(
z9$WuTOFzomo2JwOn2*2+51+Y3?OwdY(2$vH$1?l4HUEdby=v0^tl`{iuDd+V58rZ<9`ph?Rk$RHqTR^M@_j9WXr2IhbN3d!vJ_eR!n!IBN
zH;NA`EwfLwNsg%12&8HV;((?j=bhM!=bcg>z}o|DamPuD%(IRv`8tsXnNuCs`yY^4-OvCdP03q
z{2Jkz0gaAcaNFY|uf?Ab0zNN^9?D}^0gouJbtD>}_*40(amp&;Eb|XCqxWch>mH3V
z^mRfif#*0}{z)uVcnvWz2B_avwDN5%e)8F2e_#Kf
z=tP8x*1a@MpCPv-l9JwDA%h7HO5mXauPvcnaFM?7boKlRr?{|RbMziNl35li4|hD3
zJ?;`~OO*kB6sTCbd3c{GdXLK4Bfs@`0L)0UhoDf0Xtpc`rfk*UP|X`tEsSm-Q;>#B
z1N-B~)AKpro>zpRuh(ZBcEGg=YbXGQI#g&5t
z&;Rv@6m3i<3mBV^B3A!T};Z2tP|f$7MG5I2xtpYx;9
       
         eurlex
-        0.3.4
+        0.3.5
       
     
    
 
diff --git a/docs/authors.html b/docs/authors.html
index b8fc41e..6a118cd 100644
--- a/docs/authors.html
+++ b/docs/authors.html
@@ -79,7 +79,7 @@
       
       
         eurlex
-        0.3.4
+        0.3.5
       
     
    
 
diff --git a/docs/index.html b/docs/index.html
index 52ee2a8..064a425 100644
--- a/docs/index.html
+++ b/docs/index.html
@@ -38,7 +38,7 @@
       
       
         eurlex
-        0.3.4
+        0.3.5
       
     
    
 
@@ -108,9 +108,9 @@ 
    
 
  • dates <- elx_make_query("directive", include_date_transpos = TRUE) %>% elx_run_query()
    • 
 
  • ids %>% dplyr::left_join(lbs) %>% dplyr::left_join(dates)
    • 
 
-
    rather than elx_make_query("directive", include_lbs = TRUE, include_date_transpos = TRUE). This approach should make it easier to understand the returned data frame(s), especially when some variables contain missing or duplicated data.
    
+
    rather than elx_make_query("directive", include_lbs = TRUE, include_date_transpos = TRUE). This approach should make it easier to understand the returned data frame(s), especially when some variables contain missing or duplicated data. Always keep an eye on whether the work and celex columns identify rows uniquely or not.
    
 
    One of the main contributions of the SPARQL requests is that we obtain a comprehensive list of identifiers that we can subsequently use to obtain more data relating to the document in question. While the results of the SPARQL queries are useful also for webscraping (with the rvest package), the function elx_fetch_data() enables us to fire GET requests to retrieve data on documents with known identifiers (including Cellar URI). The function currently enables downloading the title and the full text of a document in all available languages.
    
-
    See the vignette for a walkthrough on how to use the package. Check function documentation for most up-to-date overview of features.
    
+
    See the vignette for a walkthrough on how to use the package. Check function documentation for most up-to-date overview of features. Example use cases are shown in this paper.
    
 
    
 
    
 
    
@@ -123,6 +123,23 @@ 
    
 
    This package nor its author are in any way affiliated with the EU Publications Office. Please refer to the applicable data reuse policies.
    
 
    Please consider contributing to the maintanance and development of the package by reporting bugs or suggesting new features.
    
 
    
+
    
+
    
+Latest changes
    
+
    
+
    
+eurlex 0.3.5
    
+
+
    
+
    
 
    
 
    
 Useful resources
    
diff --git a/docs/news/index.html b/docs/news/index.html
index e985a87..01d636d 100644
--- a/docs/news/index.html
+++ b/docs/news/index.html
@@ -79,7 +79,7 @@
       
       
         eurlex
-        0.3.4
+        0.3.5
       
     
    
 
@@ -139,7 +139,7 @@ 
    
 
      
 
    • it is now possible to select all resource types available with elx_make_query(resource_type = "any"). Since there are nearly 1 million CELEX codes, use with discretion and expect long execution times
      • 
 
    • results can be restricted to a particular directory code with elx_make_query(directory = "18") (directory code “18” denotes Common Foreign and Security Policy)
      • 
-
    • results can be restricted to a particular sector with elx_make_query(sector = 2) (sector code 3 denotes EU international agreements)
      • 
+
    • results can be restricted to a particular sector with elx_make_query(sector = 2) (sector code 2 denotes EU international agreements)
      • 
 
    
 
    
 
    
diff --git a/docs/pkgdown.yml b/docs/pkgdown.yml
index 26598c4..c90be6b 100644
--- a/docs/pkgdown.yml
+++ b/docs/pkgdown.yml
@@ -3,5 +3,5 @@ pkgdown: 1.5.1
 pkgdown_sha: ~
 articles:
   eurlexpkg: eurlexpkg.html
-last_built: 2021-03-08T23:15Z
+last_built: 2021-03-14T13:10Z
 
diff --git a/docs/reference/elx_council_votes.html b/docs/reference/elx_council_votes.html
index 5c46eb8..07882fe 100644
--- a/docs/reference/elx_council_votes.html
+++ b/docs/reference/elx_council_votes.html
@@ -80,7 +80,7 @@
       
       
         eurlex
-        0.3.4
+        0.3.5
       
     
    
 
diff --git a/docs/reference/elx_curia_list.html b/docs/reference/elx_curia_list.html
index 67f64ed..4a9e65f 100644
--- a/docs/reference/elx_curia_list.html
+++ b/docs/reference/elx_curia_list.html
@@ -81,7 +81,7 @@
       
       
         eurlex
-        0.3.4
+        0.3.5
       
     
    
 
@@ -137,7 +137,10 @@ 
    Scrape list of court cases from Curia
    
 CELEX identifiers are extracted from hyperlinks where available.
    
     
    
 
-    
    elx_curia_list(data = c("all", "ecj_old", "ecj_new", "gc_all", "cst_all"))
    
+    
    elx_curia_list(
+  data = c("all", "ecj_old", "ecj_new", "gc_all", "cst_all"),
+  parse = TRUE
+)
    
 
     
    Arguments
    
     
@@ -147,11 +150,16 @@ 
    Arg
       
    
+    
+      
+      
+    
    Data to be scraped from four separate lists of cases maintained by Curia, defaults to "all"
 which contains cases from Court of Justice, General Court and Civil Service Tribunal.
    		
     
    parse
    If `TRUE`, references to cases and appeals are parsed out from `case_info` into separate columns
    
     
    
 
     
    Value
    
 
-    
    A data frame containing case identifiers and information as character columns.
    
+    
    A data frame containing case identifiers and information as character columns. Where the case id
+contains a hyperlink to Eur-Lex, the CELEX identifier is retrieved as well.
    
 
     
    Examples
    
     
    # \donttest{
diff --git a/docs/reference/elx_fetch_data.html b/docs/reference/elx_fetch_data.html
index d61b510..cc5b191 100644
--- a/docs/reference/elx_fetch_data.html
+++ b/docs/reference/elx_fetch_data.html
@@ -80,7 +80,7 @@
       
       
         eurlex
-        0.3.4
+        0.3.5
       
     
    
 
diff --git a/docs/reference/elx_label_eurovoc.html b/docs/reference/elx_label_eurovoc.html
index eef7929..b20829e 100644
--- a/docs/reference/elx_label_eurovoc.html
+++ b/docs/reference/elx_label_eurovoc.html
@@ -80,7 +80,7 @@
       
       
         eurlex
-        0.3.4
+        0.3.5
       
     
    
 
diff --git a/docs/reference/elx_make_query.html b/docs/reference/elx_make_query.html
index f146474..12f39b1 100644
--- a/docs/reference/elx_make_query.html
+++ b/docs/reference/elx_make_query.html
@@ -82,7 +82,7 @@
       
       
         eurlex
-        0.3.4
+        0.3.5
       
     
    
 
@@ -141,8 +141,10 @@ 
    Create SPARQL quries
    
 
     
    elx_make_query(
   resource_type = c("directive", "regulation", "decision", "recommendation", "intagr",
-    "caselaw", "manual", "proposal", "national_impl"),
+    "caselaw", "manual", "proposal", "national_impl", "any"),
   manual_type = "",
+  directory = NULL,
+  sector = NULL,
   include_corrigenda = FALSE,
   include_celex = TRUE,
   include_lbs = FALSE,
@@ -150,10 +152,14 @@ 
    Create SPARQL quries
    
   include_date_force = FALSE,
   include_date_endvalid = FALSE,
   include_date_transpos = FALSE,
+  include_date_lodged = FALSE,
   include_force = FALSE,
   include_eurovoc = FALSE,
   include_author = FALSE,
   include_citations = FALSE,
+  include_court_procedure = FALSE,
+  include_directory = FALSE,
+  include_sector = FALSE,
   order = FALSE,
   limit = NULL
 )
    
@@ -169,6 +175,14 @@ 
    Arg
       manual_type
       
    Define manually the type of resource to be retrieved
    
     
+    
+      directory
+      
    Restrict the results to a given directory code
    
+    
+    
+      sector
+      
    Restrict the results to a given sector code
    
+    
     
       include_corrigenda
       
    If `TRUE`, results include corrigenda
    
@@ -197,6 +211,10 @@ 
    Arg
       include_date_transpos
       
    If `TRUE`, results include date of transposition deadline for directives
    
     
+    
+      include_date_lodged
+      
    If `TRUE`, results include date a court case was lodged with the court
    
+    
     
       include_force
       
    If `TRUE`, results include whether legislation is in force
    
@@ -213,6 +231,18 @@ 
    Arg
       include_citations
       
    If `TRUE`, results include citations (CELEX-labelled)
    
     
+    
+      include_court_procedure
+      
    If `TRUE`, results include type of court procedure and outcome
    
+    
+    
+      include_directory
+      
    If `TRUE`, results include the Eur-Lex directory code
    
+    
+    
+      include_sector
+      
    If `TRUE`, results include the Eur-Lex sector code
    
+    
     
       order
       
    Order results by ids
    
diff --git a/docs/reference/elx_parse_xml.html b/docs/reference/elx_parse_xml.html
index 1bf612e..958f88d 100644
--- a/docs/reference/elx_parse_xml.html
+++ b/docs/reference/elx_parse_xml.html
@@ -80,7 +80,7 @@
       
       
         eurlex
-        0.3.4
+        0.3.5
       
     
    
 
diff --git a/docs/reference/elx_run_query.html b/docs/reference/elx_run_query.html
index 9f0551a..53a9bc9 100644
--- a/docs/reference/elx_run_query.html
+++ b/docs/reference/elx_run_query.html
@@ -81,7 +81,7 @@
       
       
         eurlex
-        0.3.4
+        0.3.5
       
     
    
 
@@ -162,7 +162,7 @@ 
    Value
    
 
     
    Examples
    
     
    # \donttest{
-elx_run_query(elx_make_query("directive", include_force = TRUE))
    #> # A tibble: 4,316 x 4
+elx_run_query(elx_make_query("directive", include_force = TRUE))
    #> # A tibble: 4,317 x 4
 #>    work                               type                          celex  force
 #>    <chr>                              <chr>                         <chr>  <chr>
 #>  1 http://publications.europa.eu/res~ http://publications.europa.e~ 31979~ false
@@ -175,7 +175,7 @@ 
    Examp
 #>  8 http://publications.europa.eu/res~ http://publications.europa.e~ 31966~ false
 #>  9 http://publications.europa.eu/res~ http://publications.europa.e~ 31974~ false
 #> 10 http://publications.europa.eu/res~ http://publications.europa.e~ 31982~ false
-#> # ... with 4,306 more rows
    # }
+#> # ... with 4,307 more rows
    # }
 
    
   
    
   
 
diff --git a/man/elx_make_query.Rd b/man/elx_make_query.Rd
index 042c2e6..d99885b 100644
--- a/man/elx_make_query.Rd
+++ b/man/elx_make_query.Rd
@@ -2,7 +2,7 @@
 % Please edit documentation in R/elx_make_query.R
 \name{elx_make_query}
 \alias{elx_make_query}
-\title{Create SPARQL quries}
+\title{Create SPARQL queries}
 \usage{
 elx_make_query(
   resource_type = c("directive", "regulation", "decision", "recommendation", "intagr",
@@ -78,7 +78,7 @@ A character string containing the SPARQL query
 \description{
 Generates pre-defined or manual SPARQL queries to retrieve document ids from Cellar.
 List of available resource types: http://publications.europa.eu/resource/authority/resource-type .
-Note that not all resource types are compatible with the pre-defined query.
+Note that not all resource types are compatible with default parameter values.
 }
 \examples{
 elx_make_query(resource_type = "directive", include_date = TRUE, include_force = TRUE)
diff --git a/vignettes/eurlexpkg.Rmd b/vignettes/eurlexpkg.Rmd
index df6e722..05b1e3d 100644
--- a/vignettes/eurlexpkg.Rmd
+++ b/vignettes/eurlexpkg.Rmd
@@ -29,6 +29,8 @@ The `eurlex` R package attempts to significantly reduce the overhead associated
 
 The `eurlex` package currently envisions the typical use-case to consist of getting bulk information about EU legislation into R as fast as possible. The package contains three core functions to achieve that objective: `elx_make_query()` to create pre-defined or customized SPARQL queries; `elx_run_query()` to execute the pre-made or any other manually input query; and `elx_fetch_data()` to fire GET requests for certain metadata to the REST API.
 
+The package also contains largely self-explanatory functions for retrieving data on EU court cases (`elx_curia_list()`) and Council votes (`elx_council_votes()`) from outside Eur-Lex.
+
 ## `elx_make_query()`: Generate SPARQL queries
 
 The function `elx_make_query` takes as its first argument the type of resource to be retrieved from the semantic database that powers Eur-Lex (and other publications) called Cellar.
@@ -55,13 +57,13 @@ The choice of resource type is then reflected in the SPARQL query generated by t
 
 ```{r}
 query_dir %>% 
-  glue::as_glue() # for nicer printing
+  cat() # for nicer printing
 
 elx_make_query(resource_type = "caselaw") %>% 
-  glue::as_glue()
+  cat()
 
 elx_make_query(resource_type = "manual", manual_type = "SWD") %>% 
-  glue::as_glue()
+  cat()
 
 ```
 
@@ -69,21 +71,33 @@ There are various ways of querying the same information in the Cellar database d
 
 The other arguments in `elx_make_query()` relate to additional metadata to be returned. The results include by default the [CELEX number](https://eur-lex.europa.eu/content/tools/TableOfSectors/types_of_documents_in_eurlex.html) and exclude corrigenda (corrections of errors in legislation). Other data needs to be opted into. Make sure to select ones that are logically compatible (e.g. case law does not have a legal basis). More options should be added in the future.
 
-Note that availability of data for each variable has an impact on the results. The data frame returned by the query will be shrunken to the size of the variable with most missing data. It is recommended to always compare results from a desired query to a minimal query requesting only celex ids.
+Note that availability of data for each variable might have an impact on the results. The data frame returned by the query might be shrunken to the size of the variable with most missing data. It is recommended to always compare results from a desired query to a minimal query requesting only celex ids.
 
 ```{r}
 elx_make_query(resource_type = "directive", include_date = TRUE, include_force = TRUE) %>% 
-  glue::as_glue()
+  cat()
 
 # minimal query: elx_make_query(resource_type = "directive")
 
 elx_make_query(resource_type = "recommendation", include_date = TRUE, include_lbs = TRUE) %>% 
-  glue::as_glue()
+  cat()
 
 # minimal query: elx_make_query(resource_type = "recommendation")
 
 ```
 
+You can also decide to not specify any resource types, in which case all types of documents will be returned. As there are over a million documents with a CELEX identifier, this is likely not efficient for a majority of users. But since version 0.3.5 it is possible to request documents belonging to a particular ["sector"](https://eur-lex.europa.eu/content/tools/TableOfSectors/types_of_documents_in_eurlex.html) or [directory code](https://eur-lex.europa.eu/browse/directories/legislation.html).
+
+```{r}
+# request documents from directory 18 ("Common Foreign and Security Policy")
+# and sector 3 ("Legal acts")
+
+elx_make_query(resource_type = "any",
+               directory = "18",
+               sector = 3) %>% 
+  cat()
+```
+
 Now that we have a query, we are ready to run it.
 
 ## `elx_run_query()`: Execute SPARQL queries
@@ -208,7 +222,9 @@ Directives become naturally outdated with time. It might be all the more interes
 
 ```{r}
 dirs %>% 
-  ggplot(aes(x = as.Date(date), y = celex)) +
+  filter(!is.na(force)) %>% 
+  mutate(date = as.Date(date)) %>% 
+  ggplot(aes(x = date, y = celex)) +
   geom_point(aes(color = force), alpha = 0.1) +
   theme(axis.text.y = element_blank(),
         axis.line.y = element_blank(),
@@ -246,6 +262,6 @@ dirs_1970_title %>%
 
 I use term-frequency inverse-document frequency (tf-idf) to weight the importance of the words in the wordcloud. If we used pure frequencies, the wordcloud would largely consist of words conveying little meaning ("the", "and", ...).
 
-This is an extremely basic application of the `eurlex` package. Much more sophisticated methods can be used to analyse both the content and metadata of European Union legislation. If the package is useful for your research, please consider citing it.
+This is an extremely basic application of the `eurlex` package. Much more sophisticated methods can be used to analyse both the content and metadata of European Union legislation. If the package is useful for your research, please consider citing the [accompanying paper](https://www.tandfonline.com/doi/full/10.1080/2474736X.2020.1870150).^[Michal Ovádek (2021) Facilitating access to data on European Union laws, Political Research Exchange, 3:1, DOI: [10.1080/2474736X.2020.1870150](https://www.tandfonline.com/doi/full/10.1080/2474736X.2020.1870150)]