11#! /bin/sh
22if [ $# != 2 ]; then
3- echo >&2 " usage: $0 BASE_DOC_COMMIT DOC_REPO"
4- echo >&2 " creates CHANGES.html in the current directory"
5- echo >&2 " for the diffs of DOC_REPO against BASE_DOC_COMMIT"
3+ echo >&2 " Usage: $0 DIFF_TEXT DOC_REPO"
4+ echo >&2 " This script generates a CHANGES.html file in the current directory"
5+ echo >&2 " and adds anchor targets in the documents within DOC_REPO"
6+ echo >&2 " based on the diff hunks in the DIFF_TEXT file."
67 exit 1
78fi
8- BASE_DOC_COMMIT =" $1 "
9+ DIFF_TEXT =" $1 "
910DOC_REPOSITORY=" $2 "
1011
11- # Wipe out chronic diffs between old doc and new doc
12- (cd $DOC_REPOSITORY && find . -name " *.html" | xargs sed -i -e ' \;<script type="application/vnd\.jupyter\.widget-state+json">;,\;</script>; d'
1312# Create CHANGES.html
1413echo ' <html>' > CHANGES.html
1514echo ' <head>' >> CHANGES.html
@@ -19,79 +18,102 @@ echo '<script>hljs.highlightAll();</script>' >> CHANGES.html
1918cat >> CHANGES.html << EOF
2019<script>
2120document.addEventListener('DOMContentLoaded', () => {
22- const baseDocURL = 'https://sagemath.netlify.app'
21+ // This URL is hardcoded in the file .github/workflows/doc-publish.yml.
22+ // See NETLIFY_ALIAS of the "Deploy to Netlify" step.
23+ const baseDocURL = 'https://doc-develop--sagemath.netlify.app'
2324const diffSite = 'https://pianomister.github.io/diffsite'
2425const diffParagraphs = document.querySelectorAll('p.diff');
2526diffParagraphs.forEach(paragraph => {
2627 const rootURL = window.location.origin;
27- const docAnchor = paragraph.querySelector('a'); // first "a" element
28+ const docAnchor = paragraph.querySelector('a');
2829 const url = new URL(docAnchor.href);
2930 const path = url.pathname;
3031 const anchor = document.createElement('a');
3132 anchor.href = diffSite + '/?url1=' + rootURL + path + '&url2=' + baseDocURL + path;
3233 anchor.textContent = 'compare with the base';
3334 anchor.setAttribute('target', '_blank');
35+ paragraph.innerHTML += ' ';
3436 paragraph.appendChild(anchor);
35- paragraph.innerHTML += ' ' ;
36- const hunkAnchors = paragraph.querySelectorAll('a. hunk');
37- hunkAnchors.forEach( hunkAnchor => {
37+ const hunks = paragraph.parentNode.querySelectorAll('p.hunk') ;
38+ hunks.forEach( hunk => {
39+ const hunkAnchor = hunk.querySelector('a');
3840 const url = new URL(hunkAnchor.href);
3941 const path = url.pathname;
4042 const pathHash = path + url.hash.replace('#', '%23');
4143 const anchor = document.createElement('a');
4244 anchor.href = diffSite + '/?url1=' + rootURL + pathHash + '&url2=' + baseDocURL + path;
43- anchor.textContent = hunkAnchor.textContent ;
45+ anchor.textContent = 'compare with the base' ;
4446 anchor.setAttribute('target', '_blank');
45- paragraph.appendChild(anchor) ;
46- paragraph.innerHTML += ' ' ;
47+ hunk.innerHTML += ' ' ;
48+ hunk.appendChild(anchor) ;
4749 });
4850});
4951});
5052</script>
5153EOF
5254echo ' </head>' >> CHANGES.html
5355echo ' <body>' >> CHANGES.html
54- (cd $DOC_REPOSITORY && git diff $BASE_DOC_COMMIT -- " *.html" > diff.txt
5556python3 - << EOF
5657import os, re, html
57- with open('diff.txt', 'r') as f:
58+ from itertools import chain
59+ with open('$DIFF_TEXT ', 'r') as f:
5860 diff_text = f.read()
5961diff_blocks = re.split(r'^(?=diff --git)', diff_text, flags=re.MULTILINE)
6062out_blocks = []
6163for block in diff_blocks:
6264 match = re.search(r'^diff --git a/(.*) b/\1', block, flags=re.MULTILINE)
6365 if match:
64- doc = match.group(1)
65- file_path = os.path.join('$DOC_REPOSITORY ', doc )
66+ path = match.group(1)
67+ file_path = os.path.join('$DOC_REPOSITORY ', path )
6668 try:
6769 with open(file_path, 'r') as file:
6870 content = file.readlines()
6971 except FileNotFoundError:
7072 content = []
7173 count = 0
74+ hunks = []
75+ hunk_lines = []
76+ in_hunk = False
7277 for line in block.splitlines():
7378 if line.startswith('@@ -'):
79+ if hunk_lines:
80+ hunks.append('<pre><code class="language-diff">'
81+ + html.escape('\n'.join(hunk_lines)).strip()
82+ + '</code></pre>')
83+ hunk_lines = []
7484 search_result = re.search(r'@@ -(\d+),(\d+) \+(\d+),(\d+)', line)
7585 if search_result:
76- line_number = int(search_result.group(3))
77- for i in range(line_number - 1, -1, -1):
78- if content[i].startswith('<'):
86+ line_number = int(search_result.group(3)) - 1
87+ span = int(search_result.group(4))
88+ for i in chain(range(line_number, line_number + span), range(line_number - 1, -1, -1)):
89+ try:
90+ ln = content[i]
91+ except IndexError:
92+ continue
93+ for idx, char in enumerate(ln):
94+ if not char.isspace():
95+ break
96+ else:
97+ idx = len(ln)
98+ if ln.startswith('<', idx) and not ln.startswith('</', idx):
7999 count += 1
80- content[i] = f'<span id="hunk{count}" style="visibility: hidden;"></span>' + content[i]
100+ content[i] = ln[:idx] + f'<span id="hunk{count}" style="visibility: hidden;"></span>' + ln[idx:]
101+ hunks.append(f'<p class="hunk"><a href="{path}#hunk{count}" class="hunk" target="_blank">hunk #{count}</a></p>')
81102 break
103+ hunk_lines.append(line)
104+ if hunk_lines:
105+ hunks.append('<pre><code class="language-diff">'
106+ + html.escape('\n'.join(hunk_lines)).strip()
107+ + '</code></pre>')
82108 if content:
83109 with open(file_path, 'w') as file:
84110 file.writelines(content)
85- path = doc
86- hunks = ' '.join(f'<a href="{path}#hunk{i+1}" class="hunk" target="_blank">#{i + 1}</a>' for i in range(count))
87- out_blocks.append(f'<p class="diff"><a href="{path}">{doc}</a> ' + hunks + ' </p>'
88- + '\n<pre><code class="language-diff">'
89- + html.escape(block).strip() + '</code></pre>')
111+ out_blocks.append(f'<div class="diff"><p class="diff"><a href="{path}">{path}</a></p>\n' + '\n'.join(hunks) + '\n</div>')
90112output_text = '\n'.join(out_blocks)
91113with open('diff.html', 'w') as f:
92114 f.write(output_text)
93115EOF
94116cat diff.html >> CHANGES.html
95117echo ' </body>' >> CHANGES.html
96118echo ' </html>' >> CHANGES.html
97- rm diff.txt diff. html
119+ rm diff.html
0 commit comments