ebook: check html syntax

Author: entorb

.github/workflows/check-pr.yml

@@ -23,8 +23,8 @@ jobs:
           persist-credentials: false
           fetch-depth: 1 # 0 if you want to push to repo
 
-      - name: Touch requirements.txt for pip caching
-        run: touch requirements.txt
+      - name: Preparations
+        run: ln -s python-requirements.txt requirements.txt
 
       - name: Python set up
         uses: actions/setup-python@v5

.github/workflows/create-release.yml

@@ -1,5 +1,6 @@
 # This manually started workflow builds PDFs and eBooks and creates a new release with these assets.
-name: "Create Release"
+
+name: Create Release
 
 on:
   workflow_dispatch: # manual triggering
@@ -31,9 +32,6 @@ jobs:
           persist-credentials: false
           fetch-depth: 1 # 0 if you want to push to repo
 
-      - name: Touch requirements.txt for pip caching
-        run: touch requirements.txt
-
       - name: Cache LaTeX files
         uses: actions/cache@v4
         with:
@@ -48,6 +46,9 @@ jobs:
             hpmor*.xdv
           key: tex-cache
 
+      - name: Preparations
+        run: ln -s python-requirements.txt requirements.txt
+
       - name: Python set up
         uses: actions/setup-python@v5
         with:
@@ -63,21 +64,41 @@ jobs:
       - name: Install requirements
         run: sh scripts/install_requirements.sh > /dev/null
 
+      - name: Download previous hpmor.html
+        run: |
+          wget --quiet https://github.com/${{ github.repository }}/releases/download/WorkInProgress/hpmor.html -O hpmor-prev.html
+
       - name: Make PDFs
         run: sh scripts/make_pdfs.sh > /dev/null
 
       - name: Make eBooks
         run: sh scripts/make_ebooks.sh
 
+      - name: Compare to previous hpmor.html
+        run: |
+          diff -u -s hpmor-prev.html hpmor.html > hpmor-html-diff.log || :
+          rm hpmor-prev.html
+
+      - name: Upload eBooks as artifact
+        uses: actions/upload-artifact@v4
+        with:
+          name: ebooks
+          path: |
+            ./hpmor-html-diff.log
+            ./hpmor.epub
+            ./hpmor.html
+            ./hpmor.pdf
+          retention-days: 14
+
       - name: Publish release
         uses: softprops/action-gh-release@v2
         with:
           tag_name: "${{ inputs.version_number }}"
           body: "${{ inputs.version_text }}"
           prerelease: false
           files: |
-            ./hpmor*.pdf
-            ./hpmor.html
             ./hpmor.epub
-            ./hpmor.mobi
             ./hpmor.fb2
+            ./hpmor.html
+            ./hpmor.mobi
+            ./hpmor*.pdf

.github/workflows/make.yml

@@ -19,10 +19,7 @@ jobs:
           persist-credentials: false
           fetch-depth: 1 # 0 if you want to push to repo
 
-      - name: Touch requirements.txt for pip caching
-        run: touch requirements.txt
-
-      - name: Caching LaTeX files
+      - name: Cache LaTeX files
         uses: actions/cache@v4
         with:
           path: |
@@ -36,6 +33,9 @@ jobs:
             hpmor*.xdv
           key: tex-cache
 
+      - name: Preparations
+        run: ln -s python-requirements.txt requirements.txt
+
       - name: Python set up
         uses: actions/setup-python@v5
         with:
@@ -90,10 +90,10 @@ jobs:
         with:
           name: ebooks
           path: |
-            ./hpmor.pdf
-            ./hpmor.html
             ./hpmor-html-diff.log
             ./hpmor.epub
+            ./hpmor.html
+            ./hpmor.pdf
           retention-days: 14
 
       - name: Upload files to release WorkInProgress
@@ -102,9 +102,9 @@ jobs:
           tag_name: WorkInProgress
           prerelease: true
           files: |
-            ./hpmor*.pdf
+            ./hpmor-html-diff.log
             ./hpmor.epub
-            ./hpmor.mobi
             ./hpmor.fb2
             ./hpmor.html
-            ./hpmor-html-diff.log
+            ./hpmor.mobi
+            ./hpmor*.pdf

Dockerfile

@@ -9,9 +9,9 @@ ENV TZ=Europe/Berlin
 # prevent keyboard input requests in apt install
 ENV DEBIAN_FRONTEND=noninteractive
 
-# install core packages and other dependencies
+# install packages and cleanup afterwards
 RUN apt-get update && apt-get dist-upgrade -y && \
-    apt-get install -y python3 git texlive-xetex texlive-lang-greek texlive-lang-german latexmk texlive-extra-utils pandoc calibre imagemagick ghostscript && \
+    apt-get install -y python3 python3-lxml git texlive-xetex texlive-lang-greek texlive-lang-german latexmk texlive-extra-utils pandoc calibre imagemagick ghostscript && \
     apt-get clean autoclean && apt-get autoremove --yes && rm -rf /var/lib/{apt,dpkg,cache,log}/
 
 # set working directory

python-requirements.txt

@@ -0,0 +1,2 @@
+lxml
+pytest

scripts/ebook/step_6.py

@@ -8,22 +8,36 @@
 
 import os
 import re
+import sys
 from pathlib import Path
 
+from lxml import etree  # pip install lxml
+
 os.chdir(Path(__file__).parent.parent.parent)
 
 source_file = Path("tmp/hpmor-epub-5-html-unmod.html")
 target_file = Path("hpmor.html")
 
 
+def check_html(cont: str) -> None:
+    """Check html syntax."""
+    parser = etree.XMLParser(recover=False)  # Do not auto-fix errors
+    try:
+        etree.fromstring(cont, parser)  # noqa: S320
+    except etree.XMLSyntaxError as e:
+        print("HTML Error:", e)
+        sys.exit(1)
+        # raise
+
+
 def fix_ellipsis(s: str) -> str:
     """
     Fix ellipsis spacing for ebooks.
     """
     # 1. remove all spaces around ellipsis
     s = re.sub(r" *… *", "…", s)
     # 2. recreate some spaces
-    # before punctuation : no space, so governed by 1.
+    # before punctuation : no space, so covered by 1.
     # between words
     s = re.sub(r"(?<=[\w])…(?=[\w])", "… ", s)
     # after punctuation: add space
@@ -33,7 +47,7 @@ def fix_ellipsis(s: str) -> str:
     s = re.sub(r"…(?=<em>)", "… ", s)
     # before opening EN-quotes: add space
     s = re.sub(r"…(?=[“])", "… ", s)
-    # before opening DE-quotes: add space
+    # NO: before opening DE-quotes: add space
     # s = re.sub(r"…(?=[„])", "… ", s)
     return s
 
@@ -43,6 +57,8 @@ def fix_ellipsis(s: str) -> str:
 
     with source_file.open(encoding="utf-8", newline="\n") as fh_in:
         cont = fh_in.read()
+    print("checking source html")
+    check_html(cont)
 
     # remove strange leftovers from tex -> html conversion
     cont = re.sub(
@@ -55,7 +71,7 @@ def fix_ellipsis(s: str) -> str:
 
     # stray </div> leftover
     cont = re.sub(
-        r"(https://github.com/rrthomas/hpmor/</a></p>)\s+</div>",
+        r"(github.com/rrthomas/hpmor/</a></p>)\s+</div>",
         r"\1",
         cont,
         flags=re.DOTALL | re.IGNORECASE,
@@ -89,15 +105,6 @@ def fix_ellipsis(s: str) -> str:
     #     count=1,
     # )
 
-    # remove training slashes to satisfy https://validator.w3.org
-    cont = cont.replace("<br />", "<br>")
-    cont = cont.replace("<hr />", "<hr>")
-    cont = re.sub(
-        r"(<meta [^>]*) />",
-        r"\1>",
-        cont,
-    )
-
     # fix spaces around ellipsis
     cont = fix_ellipsis(cont)
 
@@ -158,5 +165,17 @@ def fix_ellipsis(s: str) -> str:
         css = fh_in.read()
     cont = cont.replace("</style>\n", css + "\n</style>\n")
 
+    print("checking target html")
+    check_html(cont)
+
+    # remove training slashes to satisfy https://validator.w3.org
+    cont = cont.replace("<br />", "<br>")
+    cont = cont.replace("<hr />", "<hr>")
+    cont = re.sub(
+        r"(<meta [^>]*) />",
+        r"\1>",
+        cont,
+    )
+
     with target_file.open(mode="w", encoding="utf-8", newline="\n") as fh_out:
         fh_out.write(cont)

scripts/install_requirements.sh

@@ -1,5 +1,9 @@
 #!/bin/sh
 
+# ensure we are in the hpmor root dir
+script_dir=$(dirname $0)
+cd $script_dir/..
+
 # update apt list
 sudo apt-get update --fix-missing
 
@@ -10,3 +14,5 @@ sudo apt-get install texlive-xetex texlive-lang-greek latexmk
 sudo apt-get install pandoc calibre texlive-extra-utils imagemagick ghostscript
 # texlive-extra-utils for latexpand
 # imagemagick ghostscript : for pdf title page to image conversion
+
+pip install -r python-requirements.txt