Merge pull request #65 from dvklopfenstein/dev

updates

README.md

@@ -1,7 +1,6 @@
 # PubMed ID (PMID) Cite
 
 [![Tweet](https://img.shields.io/twitter/url/http/shields.io.svg?style=social)](https://twitter.com/intent/tweet?text=Python%20library%20to%20download%20pubmed%20citation%20counts%20and%20data,%20given%20a%20PMID&url=https://github.com/dvklopfenstein/pmidcite&via=dvklopfenstein&hashtags=pubmed,pmid,citations,pubmed2cite,writingtips,scientificwriting)
-[![build](https://github.com/dvklopfenstein/pmidcite/actions/workflows/build.yml/badge.svg)](https://github.com/dvklopfenstein/pmidcite/actions/workflows/build.yml)
 [![CodeQL](https://github.com/dvklopfenstein/pmidcite/actions/workflows/codeql-analysis.yml/badge.svg)](https://github.com/dvklopfenstein/pmidcite/actions/workflows/codeql-analysis.yml)
 [![Latest PyPI version](https://img.shields.io/pypi/v/pmidcite.svg)](https://pypi.org/project/pmidcite/)
 [![DOI](https://zenodo.org/badge/DOI/10.5281/zenodo.5172712.svg)](https://doi.org/10.5281/zenodo.5172712)

makefile

@@ -18,11 +18,6 @@ p:
 d:
 	find src -regextype posix-extended -regex "[a-z./]*" -type d
 
-g:
-	git status -uno
-	git remote -v
-	git branch
-
 cli:
 	find src/pmidcite/cli -name \*.py
 
@@ -129,4 +124,4 @@ clobber_tmp:
 	rm -rf ./src/tests/icite
 
 clobber:
-	make -f makefile clobber_tmp clean_build
+	make -f makefile clobber_tmp clean_build pyc

setup.py

@@ -39,6 +39,10 @@ def get_long_description():
     with open(join(dir_cur, 'README.md'), 'rb') as ifstrm:
         return ifstrm.read().decode("UTF-8")
 
+CONSOLE_SCRIPTS = [
+    'icite=pmidcite.scripts.icite:main',
+    'sumpaps=pmidcite.scripts.summarize_papers:main',
+]
 
 setup(
     name=NAME,
@@ -49,10 +53,7 @@ def get_long_description():
     package_dir=PACKAGE_DIRS,
     scripts=glob('src/bin/*.py'),
     entry_points={
-        'console_scripts':[
-            'icite=pmidcite.scripts.icite:main',
-            'sumpaps=pmidcite.scripts.summarize_papers:main',
-        ],
+        'console_scripts': CONSOLE_SCRIPTS,
     },
     # https://pypi.org/classifiers/
     classifiers=[

src/pmidcite/cfg.py

@@ -185,13 +185,14 @@ def _chk_apikey(self, loaded):
         """Check to see that user has added a NCBI API key"""
         try:
             int(loaded['apikey'], 16)
-        except ValueError:
-            msg = ('SET API KEY IN {CFG}\n'
+        except ValueError as exc:
+            msg = (f'SET API KEY IN {self.cfgfile}\n'
                    'Get an NCBI API key to run the E-utilities:\n'
                    'https://ncbiinsights.ncbi.nlm.nih.gov/2017/11/02/'
                    'new-api-keys-for-the-e-utilities\n'
-                   'To ensure your API key is not made public, add {CFG} to the .gitignore')
-            raise RuntimeError(msg.format(CFG=self.cfgfile))
+                   'To ensure your API key is not made public, '
+                   f'add {self.cfgfile} to the .gitignore')
+            raise RuntimeError(msg) from exc
 
     def _err_notfound(self):
         """Report the config file was not found"""

src/pmidcite/eutils/cmds/base.py

@@ -211,48 +211,25 @@ def epost(self, database, ids, num_ids_p_epost=10):
         ret['querykey'] = rsp['querykey']
         return ret
 
-    @staticmethod
-    def _return_einforesult(record):
-        """Return EInfo result"""
-        einforesult = record['einforesult']
-        cmdtype = record['header']['type']
-        if 'dblist' in einforesult:
-            return einforesult['dblist']
-        if cmdtype == 'einfo' and 'dbinfo' in einforesult:
-            assert len(record['einforesult']['dbinfo']) == 1
-            ## print('RRRRRRRRRRRRRRR', record.keys())
-            ## print('RRRRRRRRRRRRRRR', len(record['einforesult']['dbinfo']))
-            ## print('RRRRRRRRRRRRRRR', record)
-            return record['einforesult']['dbinfo'][0]
-        raise RuntimeError('IMPLEMENT _return_einforesult')
-
-    @staticmethod
-    def _return_linksets(record):
-        """Return ELink result"""
-        links_all = []
-        for dct0 in record['linksets']:
-            ## print('DCT', dct0)
-            if 'linksetdbs' in dct0:
-                for dct1 in dct0['linksetdbs']:
-                    links_all.extend(dct1['links'])
-        print('{N} LINKED ITEMS'.format(N=len(links_all)))
-        return links_all
-
     # ------------------------------------------------------------------------------------
     def run_eutilscmd(self, cmd, **params):  # params=None, post=None, ecitmatch=False):
         """Run NCBI E-Utilities command"""
         # params example: db retstart retmax rettype retmode webenv query_key
+        # print('RUN NCBI EUTILS CMD', cmd)
         rsp_dct = self.run_req(cmd, **params) # post=None, ecitmatch=False):
-        ## print('RRRRRRRRRRRRRRRRRRRRRRR', rsp_dct)
+        # print('RRRRRRRRRRRRRRRRRRRRRRR', rsp_dct.keys())
+        # dict_keys(['code', 'msg', 'url', 'headers', 'data'])
+        # print('RRRRRRRRRRRRRRRRRRRRRRR', rsp_dct['data'])
+        # print('RRRRRRRRRRRRRRRRRRRRRRR', rsp_dct)
         if rsp_dct is not None:
             return self._extract_rsp(rsp_dct['data'], params.get('retmode'))
         return None
 
     def _mk_cgi(self, cmd, **params):
         """Get Fast Common Gateway Interface (fcgi) string, given E-utils command/parameters"""
         cgi = self.cgifmt.format(ECMD=cmd)
+        ##print('PARAMS', params)
         params = self._construct_params(params)
-        ## print('PARAMS', params)
         options = self._encode_options(params)
         cgi += '?' + options
         return cgi
@@ -333,29 +310,19 @@ def _extract_rsp(self, record, retmode):
         """Extract the data from a response from running a Entrez Utilities command"""
         if retmode == 'json':
             try:
-                dct = json.loads(record)
-                if 'esearchresult' in dct:
-                    return dct['esearchresult']
-                if 'einforesult' in dct:
-                    return self._return_einforesult(dct)
-                if 'linksets' in dct:
-                    return self._return_linksets(dct)
-                print('KEYS:', dct.keys())
-                print('DCT:', dct)
-                raise RuntimeError('UNKNOWN RESULT in _run_req')
+                return json.loads(record)
             except json.decoder.JSONDecodeError as errobj:
                 print('JSONDecodeError = {ERR}'.format(ERR=str(errobj)))
                 traceback.print_exc()
                 print('\n**FATAL JSONDecodeError:\n{RECORD}'.format(RECORD=record.decode('utf-8')))
 
-        if retmode == 'text':
+        if retmode in {'text', 'asn.1'}:
             ## print('RECORD:', str(record))
             return record.decode('utf-8')
 
         ## print('RETMODE', retmode)
         ## print('RECORD', record)
 
-        ## print(record)
         # <?xml version="1.0" encoding="ISO-8859-1"?>
         # <!DOCTYPE ePostResult
         #      SYSTEM "https://eutils.ncbi.nlm.nih.gov/eutils/dtd/20090526/epost.dtd"
@@ -366,8 +333,11 @@ def _extract_rsp(self, record, retmode):
         # </ePostResult>
         # Parse XML
         root = ElementTree.fromstring(record)
-        ## print('root.tag', root.tag)
-        assert root.tag in 'ePostResult', root.tag
+        #print(f'ElementTree.fromstring(record).root:\n{root}')
+        return root
+        # TODO
+        #print('root.tag', root.tag)
+        assert root.tag in 'ePostResult', f'ElementTree.fromstring(record).tag: {root.tag}'
         dct = {r.tag.lower():r.text for r in root}
         if 'querykey' in dct:
             dct['querykey'] = int(dct['querykey'])

src/pmidcite/eutils/cmds/elink.py

@@ -1,22 +1,21 @@
-"""Fetch items and write"""
+"""ELink"""
 
 __author__ = 'DV Klopfenstein, PhD'
 __copyright__ = "Copyright (C) 2016-present DV Klopfenstein, PhD. All rights reserved."
 __license__ = "GPL"
 
 import sys
 import re
-from pmidcite.eutils.cmds.cmdbase import CommandBase
+from pmidcite.eutils.cmds.base import EntrezUtilities
 
 
-# TBD:
-class ELink(CommandBase):
-    """Fetch and write text"""
+class ELink(EntrezUtilities):
+    """ELink"""
 
     # pylint: disable=too-many-arguments
-    def __init__(self, retmax=10000, rettype='medline', retmode='text', batch_size=100, **kws):
-        kws_base = {k:v for k, v in kws.items() if k in CommandBase.exp_kws}
-        super(ELink, self).__init__(**kws_base)
+    def __init__(self, email, apikey, tool, batch_size=100):
+        super().__init__(email, apikey, tool)
+        self.batch_size = batch_size
 
     def elink(self, database_from, linkname, webenv, querykey, num_fetches):
         """EFetch records found for PMIDs, page by page"""
@@ -29,7 +28,6 @@ def elink(self, database_from, linkname, webenv, querykey, num_fetches):
             ## sys.stdout.write(msg)
             record = None
             try:
-                # pylint: disable=bad-whitespace
                 record = self.run_eutilscmd(
                     'elink',
                     db        = database_from,
@@ -41,15 +39,15 @@ def elink(self, database_from, linkname, webenv, querykey, num_fetches):
                     query_key = querykey)
                 print('ELINK:', linkname, record)
             except IOError as err:
-                msg = "\n*FATAL: EFetching FAILED: {}".format(err)
+                msg = f"\n*FATAL: EFetching FAILED: {err}"
                 sys.stdout.write(msg)
-                sys.stdout.write("  database:   {}\n".format(database_from))
-                sys.stdout.write("  retstart:   {}\n".format(start))
-                # sys.stdout.write("  retmax:     {}\n".format(retmax))
-                sys.stdout.write("  batch_size: {}\n".format(self.batch_size))
-                sys.stdout.write("  linkname:   {}\n".format(linkname))
-                sys.stdout.write("  webenv:     {}\n".format(webenv))
-                sys.stdout.write("  querykey:   {}\n".format(querykey))
+                sys.stdout.write(f"  database:   {database_from}\n")
+                sys.stdout.write(f"  retstart:   {start}\n")
+                # sys.stdout.write(f"  retmax:     {retmax}\n")
+                sys.stdout.write(f"  batch_size: {self.batch_size}\n")
+                sys.stdout.write(f"  linkname:   {linkname}\n")
+                sys.stdout.write(f"  webenv:     {webenv}\n")
+                sys.stdout.write(f"  querykey:   {querykey}\n")
 
             if record is not None:
                 try:
@@ -61,7 +59,7 @@ def elink(self, database_from, linkname, webenv, querykey, num_fetches):
                     # ostrm.flush()
                 # pylint: disable=broad-except
                 except Exception as err:
-                    sys.stdout.write("*FATAL: BAD READ SOCKET HANDLE: {}\n".format(str(err)))
+                    sys.stdout.write(f"*FATAL: BAD READ SOCKET HANDLE: {str(err)}\n")
             else:
                 sys.stdout.write("*FATAL: NO SOCKET HANDLE TO READ FROM\n")