test: Replace PyPDF2 with pdfplumber

bin/walkpdf

@@ -1,4 +1,4 @@
-#!/usr/bin/env python
+#!/usr/bin/env python3
 # -*- coding: utf-8 -*-
 # Copyright The IETF Trust 2018, All Rights Reserved
 
@@ -16,7 +16,7 @@ for filename in sys.argv[1:]:
         print('Could not find "%s"' % filename)
     print('File: %s' % filename)
     doc = pyobj(filename)
-    with io.open(filename+'.json', 'bw') as j:
+    with io.open(filename+'.json', 'w') as j:
         json.dump(doc, j, indent=2)
     print('Wrote: %s' % j.name)
     with io.open(filename+'.xml', 'w', encoding='utf-8') as x:

configtest.py

@@ -12,7 +12,7 @@
 for (pname, mname) in [
             ('decorator', 'decorator'),
             ('dict2xml', 'dict2xml'),
-            ('PyPDF2', 'PyPDF2'),
+            ('pdfplumber', 'pdfplumber'),
         ]:
     try:
         sys.stderr.write("  '%s'...\n" % pname)

tox.ini

@@ -29,7 +29,8 @@ allowlist_externals =
 
 deps =
     -rrequirements.txt
+    typing-extensions
     decorator
-    dict2xml==1.7
-    pypdf2<1.27.0
+    pdfplumber
+    dict2xml
     weasyprint==55.0

xml2rfc/walkpdf.py

@@ -9,86 +9,36 @@
 import json
 import lxml
 import os
-import PyPDF2 as pypdf2
+import pdfplumber
 import sys
 
-def walk(obj, seen):
-    dobj = {}                            # Direct objects
-    iobj = []                            # Indirect objects
-    if hasattr(obj, 'keys'):
-        for key in obj.keys():
-            k = key[1:] if key.startswith('/') else key
-            d, i = walk(obj[key], seen)
-            dobj[k] = d
-            iobj += i
-        if hasattr(obj, 'extractText'):
-            dobj['text'] = obj.extractText()
-    elif isinstance(obj, pypdf2.generic.ArrayObject):
-        dobj = []
-        for o in obj:
-            d, i = walk(o, seen)
-            dobj.append(d)
-            iobj += i
-    elif isinstance(obj, pypdf2.generic.BooleanObject):
-        dobj = obj.value
-    elif isinstance(obj, pypdf2.generic.NameObject):
-        dobj = str(obj)
-    elif isinstance(obj, pypdf2.generic.NumberObject):
-        dobj = int(obj)
-    elif isinstance(obj, pypdf2.generic.FloatObject):
-        dobj = float(obj)
-    elif isinstance(obj, pypdf2.generic.IndirectObject):
-        dobj = str(obj)
-        if (obj.idnum, obj.generation) not in seen:
-            seen.add((obj.idnum, obj.generation))
-            d, i = walk(obj.getObject(), seen)
-            if isinstance(d, dict):
-                d['IdNum'] = obj.idnum
-                d['Generation'] = obj.generation
-            else:
-                dobj = d
-            iobj += i
-            iobj.append(d)
-    elif isinstance(obj, pypdf2.generic.TextStringObject):
-        dobj = str(obj)
-    else:
-        raise RuntimeError("Unexpected object type: %s" % type(obj))
-
-    if hasattr(obj, 'idnum'):
-        seen.add((obj.idnum, obj.generation))
-
-    return dobj, iobj
+def get_fonts(page):
+    fonts = []
+    if 'char' in page.objects.keys():
+        for obj in page.chars:
+            # pdfplumber presents font names like `ROTXYT+Noto-Sans-Cherokee`
+            fontname = obj['fontname'].split('+')[1].replace('-', ' ')
+            if fontname not in fonts:
+                fonts.append(fontname)
+    return fonts
 
 def pyobj(filename=None, bytes=None):
-    seen = set()
-    #
     pdffile = io.BytesIO(bytes) if bytes else io.open(filename, 'br')
-    reader = pypdf2.PdfFileReader(pdffile, strict=False)
-    info = reader.getDocumentInfo()
-    doc = {}
-    for key in info.keys():
-        k = key[1:] if key.startswith('/') else key
-        doc[k] = info[key]
-    iobj = []
+    reader = pdfplumber.open(pdffile)
+    doc = reader.metadata
     pages = []
-    for num in range(reader.getNumPages()):
-        page = reader.getPage(num)
-        obj = page.getObject()
-        d, i = walk(obj, seen)
-        #pages[num+1] = d
-        pages.append(d)
-        iobj += i
+    for num in range(len(reader.pages)):
+        page = reader.pages[num]
+        pages.append({
+            'text': page.extract_text(),
+            'FontFamily': get_fonts(page)})
     pdffile.close()
-    #
     doc['Page'] = pages
-    doc['IndirectObject'] = iobj
     return doc
 
 def xmltext(filename=None, obj=None, bytes=None):
     if obj is None:
         obj = pyobj(filename=filename, bytes=bytes)
-#     for i,p in enumerate(obj['Pages']):
-#         obj['Pages'][i] = {'Page': p}
     return dict2xml.dict2xml(obj, wrap="Document")    
 
 def xmldoc(filename, text=None, bytes=None):
@@ -110,4 +60,4 @@ def main():
         print('Wrote: %s' % x.name)
 
 if __name__ == "__main__":
-    main()
+    main()