From 00956b55bf7249b1d45d0cd81243eee88b21e3e3 Mon Sep 17 00:00:00 2001
From: Peter Krempa <pkrempa@redhat.com>
Date: Tue, 14 Feb 2023 12:35:23 +0100
Subject: [PATCH] scripts: check-html-references: Improve error messages and
 don't mess with relative paths
MIME-Version: 1.0
Content-Type: text/plain; charset=utf8
Content-Transfer-Encoding: 8bit

Now that we have the source file name as a custom attribute we can use
it to report which file actually needs to be edited to fix the error:

 ERROR: 'docs/uri.rst': broken link to: 'drvqemu.html#exaple'

rather than:

 broken link targets:
 docs/uri.html broken link: drvqemu.html#exaple

which pointed to file which does not exist in the source directory.

This also allows us to delete all the relative path handling needed to
report at least somewhat user-legible errors before.

Signed-off-by: Peter Krempa <pkrempa@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
---
 scripts/check-html-references.py | 46 +++++++++++++-------------------
 1 file changed, 19 insertions(+), 27 deletions(-)

diff --git a/scripts/check-html-references.py b/scripts/check-html-references.py
index 6d9116585e..8e726928e0 100755
--- a/scripts/check-html-references.py
+++ b/scripts/check-html-references.py
@@ -30,13 +30,6 @@ def get_file_list(prefix):
     filelist = []
 
     for root, dir, files in os.walk(prefix):
-        prefixbase = os.path.dirname(prefix)
-
-        if root.startswith(prefixbase):
-            relroot = root[len(prefixbase):]
-        else:
-            relroot = root
-
         for file in files:
             if not re.search('\\.html$', file):
                 continue
@@ -45,20 +38,21 @@ def get_file_list(prefix):
             if '404.html' in file:
                 continue
 
-            fullfilename = os.path.join(root, file)
-            relfilename = os.path.join(relroot, file)
-            filelist.append((fullfilename, relfilename))
+            filelist.append(os.path.join(root, file))
 
     return filelist
 
 
 # loads an XHTML and extracts all anchors, local and remote links for the one file
-def process_file(filetuple):
-    filename, relfilename = filetuple
+def process_file(filename):
     tree = ET.parse(filename)
     root = tree.getroot()
+    docname = root.get('data-sourcedoc')
 
-    anchors = [relfilename]
+    if not docname:
+        docname = filename
+
+    anchors = [filename]
     targets = []
 
     for elem in root.findall('.//html:a', ns):
@@ -66,30 +60,30 @@ def process_file(filetuple):
         an = elem.get('id')
 
         if an:
-            anchors.append(relfilename + '#' + an)
+            anchors.append(filename + '#' + an)
 
         if target:
             if re.search('://', target):
                 externallinks.append(target)
             elif target[0] != '#' and 'mailto:' not in target:
-                dirname = os.path.dirname(relfilename)
-                targetname = os.path.normpath(os.path.join(dirname, target))
+                dirname = os.path.dirname(filename)
+                targetfull = os.path.normpath(os.path.join(dirname, target))
 
-                targets.append((targetname, filename, target))
+                targets.append((filename, docname, targetfull, target))
 
     # older docutils generate "<div class='section'"
     for elem in root.findall('.//html:div/[@class=\'section\']', ns):
         an = elem.get('id')
 
         if an:
-            anchors.append(relfilename + '#' + an)
+            anchors.append(filename + '#' + an)
 
     # modern docutils generate a <section element
     for elem in root.findall('.//html:section', ns):
         an = elem.get('id')
 
         if an:
-            anchors.append(relfilename + '#' + an)
+            anchors.append(filename + '#' + an)
 
     return (anchors, targets)
 
@@ -98,8 +92,8 @@ def process_all(filelist):
     anchors = []
     targets = []
 
-    for filetuple in filelist:
-        anchor, target = process_file(filetuple)
+    for file in filelist:
+        anchor, target = process_file(file)
 
         targets = targets + target
         anchors = anchors + anchor
@@ -109,17 +103,15 @@ def process_all(filelist):
 
 def check_targets(targets, anchors):
     errors = []
-    for target, targetfrom, targetorig in targets:
+    for _, docname, target, targetorig in targets:
         if target not in anchors:
-            errors.append((targetfrom, targetorig))
+            errors.append((docname, targetorig))
 
     if errors:
         errors.sort()
 
-        print('broken link targets:')
-
         for file, target in errors:
-            print(file + " broken link: " + target)
+            print(f'ERROR: \'{file}\': broken link to: \'{target}\'')
 
         return True
 
@@ -134,7 +126,7 @@ parser.add_argument('--external', action="store_true",
 
 args = parser.parse_args()
 
-files = get_file_list(args.webroot)
+files = get_file_list(os.path.abspath(args.webroot))
 
 targets, anchors = process_all(files)
 
-- 
2.47.2