test

.github/workflows/build.yml

@@ -23,7 +23,7 @@ jobs:
 
       - run: |
           sudo apt update
-          sudo apt install -y libpoppler-glib-dev pkg-config
+          sudo apt install -y libpoppler-glib-dev pkg-config wget
 
       - uses: ankane/setup-postgres@v1
         with:

Makefile

@@ -17,11 +17,13 @@ TESTS = $(wildcard test/sql/*.sql)
 REGRESS = $(patsubst test/sql/%.sql,%,$(TESTS))
 REGRESS_OPTS = --inputdir=test --load-extension=$(EXTENSION)
 
-TEST_FILES = /tmp/pgintro.pdf /tmp/bad.pdf
+TEST_FILES = /tmp/pgintro.pdf /tmp/bad.pdf /tmp/big.pdf
 /tmp/pgintro.pdf:
 	cp test/pgintro.pdf $@
 /tmp/bad.pdf:
 	echo 'not a pdf' >> $@
+/tmp/big.pdf:
+	wget https://www.postgresql.org/files/documentation/pdf/17/postgresql-17-A4.pdf -O $@
 
 installcheck: $(TEST_FILES)
 

sql/pgpdf--0.1.0.sql

@@ -18,7 +18,8 @@ CREATE TYPE pdf
 (
     INTERNALLENGTH = -1,
     INPUT = pdf_in,
-    OUTPUT = pdf_out
+    OUTPUT = pdf_out,
+    STORAGE = extended
 );
 
 CREATE CAST (pdf AS text) WITH INOUT AS ASSIGNMENT;

test/expected/pgpdf.out

@@ -405,3 +405,14 @@ select pdf_read_bytes(pg_read_binary_file('/tmp/pgintro.pdf')) = '/tmp/pgintro.p
  t
 (1 row)
 
+/* bigger files: >8KB */
+CREATE TABLE pdfs(i serial primary key, d pdf);
+INSERT INTO pdfs(d) VALUES('/tmp/pgintro.pdf');
+INSERT INTO pdfs(d) VALUES('/tmp/big.pdf');
+SELECT length(d::text) FROM pdfs
+ length  
+---------
+    4182
+ 7055974
+(2 rows)
+

test/sql/pgpdf.sql

@@ -29,6 +29,11 @@ SELECT '/tmp/pgintro.pdf'::pdf::text @@ to_tsquery('postgres');
 SELECT '/tmp/pgintro.pdf'::pdf::text @@ to_tsquery('oracle');
 
 /* Old functions mentioned in the blog post. Making sure they continue to work */
-
 SELECT pdf_read_file('/tmp/pgintro.pdf') = '/tmp/pgintro.pdf'::pdf::text;
-select pdf_read_bytes(pg_read_binary_file('/tmp/pgintro.pdf')) = '/tmp/pgintro.pdf'::pdf::text;
+select pdf_read_bytes(pg_read_binary_file('/tmp/pgintro.pdf')) = '/tmp/pgintro.pdf'::pdf::text;
+
+/* bigger files: >8KB */
+CREATE TABLE pdfs(i serial primary key, d pdf);
+INSERT INTO pdfs(d) VALUES('/tmp/pgintro.pdf');
+INSERT INTO pdfs(d) VALUES('/tmp/big.pdf');
+SELECT length(d::text) FROM pdfs