feat: Add option to update datatracker references

With new commandline option `--use-bib`, xinclude with
datatracker.ietf.org are replaced with bib.ietf.org.

Fixes ietf-tools#1167

xml2rfc/__init__.py

@@ -16,7 +16,7 @@
 from xml2rfc.writers import ( BaseRfcWriter, RawTextRfcWriter, PaginatedTextRfcWriter,
 	 HtmlRfcWriter, NroffRfcWriter, ExpandedXmlWriter, RfcWriterError,
          V2v3XmlWriter, PrepToolWriter, TextWriter, HtmlWriter, PdfWriter,
-         ExpandV3XmlWriter, UnPrepWriter, DocWriter
+         ExpandV3XmlWriter, UnPrepWriter, DocWriter, DatatrackerToBibConverter,
      )
 
 # This defines what 'from xml2rfc import *' actually imports:
@@ -25,7 +25,7 @@
            'HtmlRfcWriter', 'NroffRfcWriter', 'ExpandedXmlWriter',
            'RfcWriterError', 'V2v3XmlWriter', 'PrepToolWriter', 'TextWriter',
            'HtmlWriter', 'PdfWriter', 'ExpandV3XmlWriter', 'UnPrepWriter', 
-           'DocWriter',
+           'DocWriter', 'DatatrackerToBibConverter',
        ]
 
 try:

xml2rfc/run.py

@@ -198,6 +198,8 @@ def main():
                            help='outputs formatted text to file, unpaginated (only v2 input)')
     formatgroup.add_argument('--expand', action='store_true',
                            help='outputs XML to file with all references expanded')
+    formatgroup.add_argument('--use-bib', action='store_true',
+                           help='update all datatracker references with bib.ietf.org')
     formatgroup.add_argument('--v2v3', action='store_true',
                            help='convert vocabulary version 2 XML to version 3')
     formatgroup.add_argument('--preptool', action='store_true',
@@ -456,7 +458,7 @@ def main():
             options.output_path = options.basename
             options.basename = None
     #
-    num_formats = len([ o for o in [options.raw, options.text, options.nroff, options.html, options.expand, options.v2v3, options.preptool, options.info, options.pdf, options.unprep ] if o])
+    num_formats = len([ o for o in [options.raw, options.text, options.nroff, options.html, options.expand, options.use_bib, options.v2v3, options.preptool, options.info, options.pdf, options.unprep ] if o])
     if num_formats > 1 and (options.filename or options.output_filename):
         sys.exit('Cannot use an explicit output filename when generating more than one format, '
                  'use --path instead.')
@@ -662,6 +664,15 @@ def main():
             options.output_filename = None
 
         # --- End of legacy formatter invocations ---
+        if options.use_bib:
+            xmlrfc = parser.parse(remove_comments=False, quiet=True, normalize=False, strip_cdata=False, add_xmlns=True)
+            filename = options.output_filename
+            if not filename:
+                filename = basename + '.bib.xml'
+                options.output_filename = filename
+            expander = xml2rfc.DatatrackerToBibConverter(xmlrfc, options=options, date=options.date)
+            expander.write(filename)
+            options.output_filename = None
 
         if options.expand and not options.legacy:
             xmlrfc = parser.parse(remove_comments=False, quiet=True, normalize=False, strip_cdata=False, add_xmlns=True)

xml2rfc/writers/__init__.py

@@ -14,11 +14,12 @@
 from xml2rfc.writers.pdf import PdfWriter
 from xml2rfc.writers.unprep import UnPrepWriter
 from xml2rfc.writers.doc import DocWriter
+from xml2rfc.writers.bib import DatatrackerToBibConverter
 
 # This defines what 'from xml2rfc.writers import *' actually imports:
 __all__ = ['BaseRfcWriter', 'RawTextRfcWriter', 'PaginatedTextRfcWriter',
            'HtmlRfcWriter', 'NroffRfcWriter', 'ExpandedXmlWriter',
            'RfcWriterError', 'V2v3XmlWriter', 'PrepToolWriter', 'TextWriter',
            'HtmlWriter', 'PdfWriter', 'ExpandV3XmlWriter', 'UnPrepWriter', 
-           'DocWriter',
+           'DocWriter', 'DatatrackerToBibConverter',
        ]

xml2rfc/writers/base.py

@@ -117,6 +117,7 @@
         'template_dir': os.path.join(os.path.dirname(os.path.dirname(__file__)), 'templates'),
         'text': True,
         'unprep': False,
+        'use_bib': False,
         'utf8': False,
         'values': False,
         'verbose': False,

xml2rfc/writers/bib.py

@@ -0,0 +1,43 @@
+# Copyright The IETF Trust 2024, All Rights Reserved
+# -*- coding: utf-8 -*-
+from __future__ import unicode_literals, print_function, division
+
+from io import open
+from lxml import etree
+from urllib.parse import urlparse
+
+from xml2rfc.writers.preptool import PrepToolWriter
+
+
+class DatatrackerToBibConverter(PrepToolWriter):
+    """Writes a duplicate XML file but with datratracker references replaced with bib.ietf.org"""
+
+    def write(self, filename):
+        """Public method to write the XML document to a file"""
+        self.convert()
+        with open(filename, "w", encoding="utf-8") as file:
+            text = etree.tostring(self.tree, encoding="unicode")
+            file.write("<?xml version='1.0' encoding='utf-8'?>\n")
+            file.write(text)
+            if not self.options.quiet:
+                self.log(" Created file %s" % filename)
+
+    def convert(self):
+        version = self.root.get("version", "3")
+        if version not in [
+            "3",
+        ]:
+            self.die(self.root, 'Expected <rfc> version="3", but found "%s"' % version)
+        self.convert_xincludes()
+
+    def convert_xincludes(self):
+        ns = {"xi": b"http://www.w3.org/2001/XInclude"}
+        xincludes = self.root.xpath("//xi:include", namespaces=ns)
+        for xinclude in xincludes:
+            href = urlparse(xinclude.get("href"))
+
+            if href.netloc == "datatracker.ietf.org":
+                reference_file = href.path.split("/")[-1]
+                xinclude.set(
+                    "href", f"https://bib.ietf.org/public/rfc/bibxml-ids/{reference_file}"
+                )