Merge pull request #3376 from eseiler/doc/linkcheck

eseiler · web-flow · commit 07e78660e846 · 2025-08-27T14:41:03.000+02:00
[DOC] Link check and doxygen-web-tags update
diff --git a/doc/about/api/index.md b/doc/about/api/index.md
@@ -92,10 +92,10 @@ this should not be a problem.
 We promise to support the above compilers in the latest release of SeqAn3, or until all the following
 operating systems provide a newer supported compiler:
 
-| Operating System             | Supported Releases¹                    |
+| Operating System             | Supported Releases                     |
 |------------------------------|----------------------------------------|
-| RedHat Enterprise Linux      | the latest release ²                   |
-| CentOS Linux                 | the latest release ²                   |
+| RedHat Enterprise Linux      | the latest release ¹                   |
+| CentOS Linux                 | the latest release ¹                   |
 | SUSE Linux Enterprise Server | the latest release                     |
 | Debian GNU/Linux             | "stable" and "old-stable"              |
 | Ubuntu Linux                 | the two latest LTS releases            |
@@ -115,9 +115,7 @@ More platforms and compilers will be added during the SeqAn3 lifecycle, but some
 **We promise to provide good forward-compatibility with the C++ standard.** And we will strive to fix any warnings that
 are added by newer versions of a supported compiler.
 
-<small>¹ [This site](https://linuxlifecycle.com) provides a good overview of what the current release and its
-lifecycle is.</small><br>
-<small>² We consider CentOS 7 / RedHat Enterprise Linux (RHEL) 7 as being community-supported. That means issues and
+<small>¹ We consider CentOS 7 / RedHat Enterprise Linux (RHEL) 7 as being community-supported. That means issues and
 patches are welcome, but we do not actively test for those operating systems. See this related
 [issue](https://github.com/seqan/seqan3/issues/2244).</small>
 
diff --git a/doc/tutorial/08_pairwise_alignment/index.md b/doc/tutorial/08_pairwise_alignment/index.md
@@ -296,7 +296,7 @@ How does the result change?
 A special form of the pairwise sequence alignment is the edit distance. This distance metric counts the number of
 edits necessary to transform one sequence into the other. The cost model for the edit distance is fixed. In particular,
 the match score is `0` and the scores for a mismatch and a gap is `-1`. Due to the special metric, a fast
-[bitvector](https://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.332.9395&rep=rep1&type=pdf) implementation can be
+[bitvector](https://doi.org/10.1145/316542.316550) implementation can be
 used to compute the edit distance. This happens in SeqAn automatically if the respective configurations are used.
 To do so, you need to explicitly use the `seqan3::hamming_scoring_scheme` as well as the default gap scheme
 initialised with `-1` for a gap and `0` for a gap open score while computing a `seqan3::global_alignment`.
diff --git a/test/documentation/seqan3-doxygen.cmake b/test/documentation/seqan3-doxygen.cmake
@@ -49,12 +49,13 @@ if (NOT EXISTS "${SEQAN3_DOXYGEN_STD_TAGFILE}" OR SEQAN3_DOXYGEN_STD_TAGFILE STR
     # When updating, check whether warnings in SEQAN3_TEST_DOXYGEN_FAIL_ON_WARNINGS are gone when removing sed filter.
     ExternalProject_Add (
         download-cppreference-doxygen-web-tag
-        URL "https://github.com/PeterFeicht/cppreference-doc/releases/download/v20241110/html-book-20241110.tar.xz"
-        URL_HASH SHA256=431e80862eb70fd4793a60d7d3b6c13c8605284978f9ea0529572e8fd1562cc6
+        URL "https://github.com/PeterFeicht/cppreference-doc/releases/download/v20250209/html-book-20250209.tar.xz"
+        URL_HASH SHA256=ac50671a1f52d7f0ab0911d14450eb35e8c2f812edd0e426b2cd1e3d9db91d6f
         TLS_VERIFY ON
         DOWNLOAD_DIR "${PROJECT_BINARY_DIR}"
         DOWNLOAD_NAME "html-book.tar.xz"
         DOWNLOAD_NO_EXTRACT YES
+        DOWNLOAD_NO_PROGRESS YES
         BINARY_DIR "${PROJECT_BINARY_DIR}"
         BUILD_BYPRODUCTS "${SEQAN3_DEFAULT_DOXYGEN_STD_TAGFILE}"
         CONFIGURE_COMMAND /bin/sh -c "xzcat html-book.tar.xz | tar -xf - cppreference-doxygen-web.tag.xml"
diff --git a/test/scripts/link_check.sh b/test/scripts/link_check.sh
@@ -1,9 +1,8 @@
 #!/usr/bin/env bash
-
 # SPDX-FileCopyrightText: 2006-2025 Knut Reinert & Freie Universität Berlin
 # SPDX-FileCopyrightText: 2016-2025 Knut Reinert & MPI für molekulare Genetik
 # SPDX-License-Identifier: BSD-3-Clause
-
+#
 # Usage: link_check.sh <SeqAn3 root directory>
 # Will output the status of links in the repository.
 #
@@ -13,27 +12,30 @@
 # The general workflow is to first run the script and then check the non-working links by searching the occurrence
 # within the codebase and verifying that they are indeed broken.
 
+COUNT=0
 do_check ()
 {
-    RESPONSE=$(curl --http2 -Is -A 'Mozilla/5.0' $1) # HTTP2 is the default.
-    if ! [[ "$RESPONSE" =~ ^HTTP.* ]]; then # If this does not work,
-        RESPONSE=$(curl --http1.1 -Is -A 'Mozilla/5.0' $1) # fall back to HTTP1.1.
+    (( COUNT++ ))
+    RESPONSE=$(curl --http2 -Is -A 'Mozilla/5.0' "$1") # HTTP2 is the default.
+    if ! [[ "${RESPONSE}" =~ ^HTTP.* ]]; then # If this does not work,
+        RESPONSE=$(curl --http1.1 -Is -A 'Mozilla/5.0' "$1") # fall back to HTTP1.1.
     fi
 
-    HEADER=($(echo $RESPONSE | head -1)); # May look like: HTTP/2 200
+    HEADER=($(echo "${RESPONSE}" | head -1)); # May look like: HTTP/2 200
     STATUS=${HEADER[1]}
+    echo -n "[${COUNT}] "
     case "$STATUS" in
-        200) echo "Link OK         :" $1;;
-        301) echo "Link PERM MOVED :" $1;;
-        302) echo "Link TEMP MOVED :" $1;;
-        404) echo "Link BROKE      :" $1;;
-        429) sleep 5; do_check $1;;
-        *)   echo "Link STATUS" $STATUS ":" $1;;
+        200) echo "Link OK         : $1";;
+        301) echo "Link PERM MOVED : $1";;
+        302) echo "Link TEMP MOVED : $1";;
+        404) echo "Link BROKE      : $1";;
+        429) sleep 5; do_check "$1";;
+        *)   echo "Link STATUS ${STATUS} : $1";;
     esac
 }
 
-if [[ $# -ne 1 ]]; then
-    echo "Usage: link_check.sh <SeqAn3 root directory>"
+if [[ $# -ne 1 ]] && [[ $# -ne 2 ]]; then
+    echo "Usage: link_check.sh <SeqAn3 root directory> <Skip first n=0 entries>"
     exit 1
 fi
 
@@ -48,7 +50,10 @@ if [[ ! -f $1/include/seqan3/version.hpp ]]; then
     exit 1
 fi
 
-for URL in $(grep -ohr --exclude-dir={.git,submodules} "https*://[a-zA-Z0-9./#+?=_%:-]*[a-zA-Z0-9/#+?=_%:-]" $1 | sort | uniq)
+SKIP="${2:-0}"
+COUNT=$((COUNT + SKIP))
+
+for URL in $(grep -ohr --exclude-dir={.git,.vscode,build} "https*://[a-zA-Z0-9./#+?=_%:-]*[a-zA-Z0-9/#+?=_%:-]" "$1" | sort | uniq | tail -n +"${SKIP}")
 do
-  do_check $URL
+  do_check "${URL}"
 done
