llms-txt - do not process external links (#14183)

Author: cscheid

src/resources/filters/llms/llms.lua

@@ -190,7 +190,8 @@ function Link(link)
         return link.content
     end
 
-    if link.target and (link.target:match("%.html$") or link.target:match("%.html#")) then
+    local is_absolute = link.target:match("^%a[%w+%-%.]*:") or link.target:match("^//")
+    if link.target and not is_absolute and (link.target:match("%.html$") or link.target:match("%.html#")) then
         link.target = link.target:gsub("%.html#", ".llms.md#")
         link.target = link.target:gsub("%.html$", ".llms.md")
         link.target = link.target:gsub("^%./", "")

tests/docs/smoke-all/website/llms-txt/about.qmd

@@ -6,9 +6,9 @@ _quarto:
       ensureLlmsMdExists: true
       ensureLlmsMdRegexMatches:
         # First array: patterns that MUST match
-        - ["^# About", "> \\*\\*NOTE:\\*\\*", "> \\*\\*WARNING:\\*\\*", "This is a note", "``` python", "def hello", "\\| Feature", "\\|[-]+\\|", "\\[home page\\]\\(.*\\.llms\\.md\\)", "\\[test site intro\\]\\(index\\.llms\\.md#test-content\\)", "## Alpha Tab", "Alpha content here", "## Beta Tab", "Beta content here"]
-        # Second array: patterns that must NOT match (no .html links, no breadcrumbs, no empty tab links)
-        - ["\\.html\\)", "\\.html#", "\\[Info\\]", "\\[Alpha Tab\\]\\(\\)", "\\[Beta Tab\\]\\(\\)"]
+        - ["^# About", "> \\*\\*NOTE:\\*\\*", "> \\*\\*WARNING:\\*\\*", "This is a note", "``` python", "def hello", "\\| Feature", "\\|[-]+\\|", "\\[home page\\]\\(.*\\.llms\\.md\\)", "\\[test site intro\\]\\(index\\.llms\\.md#test-content\\)", "## Alpha Tab", "Alpha content here", "## Beta Tab", "Beta content here", "https://pandoc.org/lua-filters.html", "https://www.lua.org/manual/5.3/manual.html#6.4"]
+        # Second array: patterns that must NOT match (no internal .html links, no breadcrumbs, no empty tab links)
+        - ["\\(index\\.html\\)", "\\(index\\.html#", "\\[Info\\]", "\\[Alpha Tab\\]\\(\\)", "\\[Beta Tab\\]\\(\\)"]
 ---
 
 About this test site.
@@ -58,3 +58,9 @@ Beta content here.
 Go back to the [home page](index.qmd).
 
 Go to the [test site intro](index.qmd#test-content).
+
+## External Link Example
+
+Check out [Pandoc filters](https://pandoc.org/lua-filters.html) for more info.
+
+See the [Lua manual](https://www.lua.org/manual/5.3/manual.html#6.4) for string functions.

tests/docs/smoke-all/website/llms-txt/index.qmd

@@ -14,8 +14,8 @@ _quarto:
       ensureLlmsMdRegexMatches:
         # First array: patterns that MUST match - verify anchor links, code annotations, and conditional content
         - ["\\[callout examples\\]\\(about\\.llms\\.md#callout-examples\\)", "# <1>", "# <2>", "Load tidyverse", "Open help for ggplot", "only for LLM consumption"]
-        # Second array: patterns that must NOT match (no .html links, no annotation UI, no hidden content)
-        - ["\\.html\\)", "\\.html#", "code-annotation-anchor", "should not appear in LLM output"]
+        # Second array: patterns that must NOT match (no internal .html links, no annotation UI, no hidden content)
+        - ["\\(about\\.html\\)", "\\(about\\.html#", "code-annotation-anchor", "should not appear in LLM output"]
 ---
 
 ## Test Content