Merge pull request #79 from jimustafa/update-checks

update checks

.github/workflows/main.yaml

@@ -51,13 +51,9 @@ jobs:
           cp -r fonts/ /usr/share/fonts/
           fc-cache
           make all
-      - name: Check layouts
+      - name: Run checks
         run: |
-          set -x
-          [[ "$(pdfinfo cheatsheets.pdf          | grep Pages | awk '{print $2}')" == "2" ]] || exit 1
-          [[ "$(pdfinfo handout-tips.pdf         | grep Pages | awk '{print $2}')" == "1" ]] || exit 1
-          [[ "$(pdfinfo handout-beginner.pdf     | grep Pages | awk '{print $2}')" == "1" ]] || exit 1
-          [[ "$(pdfinfo handout-intermediate.pdf | grep Pages | awk '{print $2}')" == "1" ]] || exit 1
+          make check
       - uses: actions/upload-artifact@v2
         if: ${{ always() }}
         with:

Makefile

@@ -36,6 +36,14 @@ handouts:
 	convert -density 150 handout-beginner.pdf handout-beginner.png
 	convert -density 150 handout-intermediate.pdf handout-intermediate.png
 
+.PHONY: check
+check:
+	./check-num-pages.sh cheatsheets.pdf 2
+	./check-num-pages.sh handout-tips.pdf 1
+	./check-num-pages.sh handout-beginner.pdf 1
+	./check-num-pages.sh handout-intermediate.pdf 1
+	./check-links.py cheatsheets.pdf
+
 .PHONY: fonts
 fonts:
 	make -C fonts/

check-links.py

@@ -0,0 +1,20 @@
+#!/usr/bin/env python
+import sys
+
+import pdfx
+
+
+pdf = pdfx.PDFx(sys.argv[1])
+
+refs = [ref for ref in pdf.get_references() if ref.reftype == 'url']
+
+status_codes = list(map(lambda ref: pdfx.downloader.get_status_code(ref.ref), refs))
+
+broken_links = [refs[idx].ref for idx in range(len(refs)) if status_codes[idx] != 200]
+
+# it seems that Twitter does not respond well to the link checker and throws a 400
+if all(['twitter.com' in url for url in broken_links]):
+    sys.exit(0)
+else:
+    print('Broken links:', broken_links)
+    sys.exit(1)

check-num-pages.sh

@@ -0,0 +1,10 @@
+#!/bin/bash
+#
+# Check that a given pdf has a certain number of pages.
+# Usage:
+#   check-num-pages.sh [pdffile] [num_pages]
+
+set -x
+pdffile=$1
+num_pages=$2
+[[ "$(pdfinfo $pdffile | grep Pages | awk '{print $2}')" == "$num_pages" ]] || exit 1

requirements/requirements.in

@@ -1,6 +1,7 @@
 bump2version
 cartopy==0.19.0.post1
 matplotlib==3.4.2
+pdfx
 pip-tools
 pre-commit
 scipy

requirements/requirements.txt

@@ -12,10 +12,18 @@ bump2version==1.0.1
     # via -r requirements.in
 cartopy==0.19.0.post1
     # via -r requirements.in
+cffi==1.15.0
+    # via cryptography
 cfgv==3.2.0
     # via pre-commit
+chardet==4.0.0
+    # via
+    #   pdfminer.six
+    #   pdfx
 click==8.0.0
     # via pip-tools
+cryptography==35.0.0
+    # via pdfminer.six
 cycler==0.10.0
     # via matplotlib
 distlib==0.3.1
@@ -40,6 +48,10 @@ numpy==1.19.5
     #   cartopy
     #   matplotlib
     #   scipy
+pdfminer.six==20201018
+    # via pdfx
+pdfx==1.4.1
+    # via -r requirements.in
 pep517==0.10.0
     # via pip-tools
 pillow==8.1.2
@@ -48,6 +60,8 @@ pip-tools==6.1.0
     # via -r requirements.in
 pre-commit==2.11.1
     # via -r requirements.in
+pycparser==2.20
+    # via cffi
 pyparsing==2.4.7
     # via matplotlib
 pyshp==2.1.3
@@ -65,6 +79,8 @@ six==1.15.0
     #   cycler
     #   python-dateutil
     #   virtualenv
+sortedcontainers==2.4.0
+    # via pdfminer.six
 toml==0.10.2
     # via
     #   pep517