Merge pull request #11921 from quarto-dev/lua/latex-table-pattern

Author: cderv

news/changelog-1.7.md

@@ -37,6 +37,7 @@ All changes included in 1.7:
 
 - ([#11835](https://github.com/quarto-dev/quarto-cli/issues/11835)): Take markdown structure into account when detecting minimum heading level.
 - ([#11903](https://github.com/quarto-dev/quarto-cli/issues/11903)): `crossref` configuration like `fig-title` or `tbl-title` now correctly supports multi word values, e.g. `fig-title: 'Supplementary Figure'`.
+- ([#11878](https://github.com/quarto-dev/quarto-cli/issues/11878)): Correctly fixup raw LaTeX table having an unexpected table env with options (e.g `\begin{table}[c]`) to be handled as crossref table.
 
 ## `typst` format
 

src/resources/filters/common/tables.lua

@@ -130,15 +130,12 @@ end
 
 function hasRawLatexTable(raw)
   if _quarto.format.isRawLatex(raw) and _quarto.format.isLatexOutput() then
-    for i,pattern in ipairs(_quarto.patterns.latexTablePatterns) do
-      if _quarto.modules.patterns.match_all_in_table(pattern)(raw.text) then
-        return true
-      end
+    local matched, _ = _quarto.modules.patterns.match_in_list_of_patterns(raw.text, _quarto.patterns.latexAllTableEnvPatterns)
+    if matched then
+      return true
     end
-    return false
-  else
-    return false
   end
+  return false
 end
 
 local tableCheckers = {

src/resources/filters/customnodes/floatreftarget.lua

@@ -448,81 +448,84 @@ end, function(float)
     local made_fix = false
     local function fix_raw(is_star_env)
       local function set_raw(el)
-        if _quarto.format.isRawLatex(el) and _quarto.modules.patterns.match_all_in_table(_quarto.patterns.latexLongtablePattern)(el.text) then
-          made_fix = true
-          local raw = el
-          -- special case for longtable floats in LaTeX
-          local extended_pattern = {".-"}
-          for _, pattern in ipairs(_quarto.patterns.latexLongtablePattern) do
-            table.insert(extended_pattern, pattern)
-          end
-          table.insert(extended_pattern, ".*")
-          local longtable_preamble, longtable_begin, longtable_content, longtable_end, longtable_postamble = _quarto.modules.patterns.match_all_in_table(extended_pattern)(raw.text)
-          if longtable_preamble == nil or longtable_begin == nil or longtable_content == nil or longtable_end == nil or longtable_postamble == nil then
-            warn("Could not parse longtable parameters. This could happen because the longtable parameters\n" ..
-            "are not well-formed or because of a bug in quarto. Please consider filing a bug report at\n" ..
-            "https://github.com/quarto-dev/quarto-cli/issues/, and make sure to include the document that\n" ..
-            "triggered this error.")
-            return {}
-          end
-          -- split the content into params and actual content
-          -- params are everything in the first line of longtable_content
-          -- actual content is everything else
-          local start, content = split_longtable_start(longtable_begin .. longtable_content)
-          if start == nil or content == nil then
-            warn("Could not parse longtable parameters. This could happen because the longtable parameters\n" ..
-            "are not well-formed or because of a bug in quarto. Please consider filing a bug report at\n" ..
-            "https://github.com/quarto-dev/quarto-cli/issues/, and make sure to include the document that\n" ..
-            "triggered this error.")
-            return {}
-          end
-          local cap_loc = cap_location(float)
-          if float.parent_id then
-            -- need to fixup subtables because longtables don't support subcaptions,
-            -- and longtable captions increment the wrong counter
-            -- we try our best here
-
-            fatal("longtables are not supported in subtables.\n" ..
-              "This is not a Quarto bug - the LaTeX longtable environment doesn't support subcaptions.\n")
-            return {}
-          end
-          if is_star_env then
-            -- content: table payload
-            -- start: \\begin{longtable}... command
-            -- longtable_preamble: everything that came before the \\begin{longtable} command
-            -- longtable_postamble: everything that came after the \\end{longtable} command
-            local result = pandoc.Blocks({
-              pandoc.RawBlock("latex", longtable_preamble),
-              pandoc.RawBlock("latex", "\\begin{table*}"),
-              -- caption here if cap_loc == "top"
-              pandoc.RawBlock("latex", start .. "\n" .. content .. "\n\\end{longtable}"),
-              -- caption here if cap_loc ~= "top"
-              pandoc.RawBlock("latex", "\\end{table*}"),
-              pandoc.RawBlock("latex", longtable_postamble),
-            })
-            if cap_loc == "top" then
-              result:insert(3, latex_caption)
-              -- gets around the padding that longtable* adds
-              result:insert(4, pandoc.RawBlock("latex", "\\vspace{-1em}"))
-            else
-              result:insert(4, latex_caption)
+        if _quarto.format.isRawLatex(el) then
+          local longtable_match, longtable_pattern = _quarto.modules.patterns.match_in_list_of_patterns(el.text, _quarto.patterns.latexLongtableEnvPatterns)
+          if longtable_match and longtable_pattern then
+            made_fix = true
+            local raw = el
+            -- special case for longtable floats in LaTeX
+            local extended_pattern = {".-"}
+            for _, pattern in ipairs(longtable_pattern) do
+              table.insert(extended_pattern, pattern)
             end
-            return result
-          else
-            local result = pandoc.Blocks({latex_caption, pandoc.RawInline("latex", "\\tabularnewline")})
-            -- if cap_loc is top, insert content on bottom
-            if cap_loc == "top" then
-              result:insert(pandoc.RawBlock("latex", content))        
+            table.insert(extended_pattern, ".*")
+            local longtable_preamble, longtable_begin, longtable_content, longtable_end, longtable_postamble = _quarto.modules.patterns.match_all_in_table(extended_pattern)(raw.text)
+            if longtable_preamble == nil or longtable_begin == nil or longtable_content == nil or longtable_end == nil or longtable_postamble == nil then
+              warn("Could not parse longtable parameters. This could happen because the longtable parameters\n" ..
+              "are not well-formed or because of a bug in quarto. Please consider filing a bug report at\n" ..
+              "https://github.com/quarto-dev/quarto-cli/issues/, and make sure to include the document that\n" ..
+              "triggered this error.")
+              return {}
+            end
+            -- split the content into params and actual content
+            -- params are everything in the first line of longtable_content
+            -- actual content is everything else
+            local start, content = split_longtable_start(longtable_begin .. longtable_content)
+            if start == nil or content == nil then
+              warn("Could not parse longtable parameters. This could happen because the longtable parameters\n" ..
+              "are not well-formed or because of a bug in quarto. Please consider filing a bug report at\n" ..
+              "https://github.com/quarto-dev/quarto-cli/issues/, and make sure to include the document that\n" ..
+              "triggered this error.")
+              return {}
+            end
+            local cap_loc = cap_location(float)
+            if float.parent_id then
+              -- need to fixup subtables because longtables don't support subcaptions,
+              -- and longtable captions increment the wrong counter
+              -- we try our best here
+
+              fatal("longtables are not supported in subtables.\n" ..
+                "This is not a Quarto bug - the LaTeX longtable environment doesn't support subcaptions.\n")
+              return {}
+            end
+            if is_star_env then
+              -- content: table payload
+              -- start: \\begin{longtable}... command
+              -- longtable_preamble: everything that came before the \\begin{longtable} command
+              -- longtable_postamble: everything that came after the \\end{longtable} command
+              local result = pandoc.Blocks({
+                pandoc.RawBlock("latex", longtable_preamble),
+                pandoc.RawBlock("latex", "\\begin{table*}"),
+                -- caption here if cap_loc == "top"
+                pandoc.RawBlock("latex", start .. "\n" .. content .. "\n\\end{longtable}"),
+                -- caption here if cap_loc ~= "top"
+                pandoc.RawBlock("latex", "\\end{table*}"),
+                pandoc.RawBlock("latex", longtable_postamble),
+              })
+              if cap_loc == "top" then
+                result:insert(3, latex_caption)
+                -- gets around the padding that longtable* adds
+                result:insert(4, pandoc.RawBlock("latex", "\\vspace{-1em}"))
+              else
+                result:insert(4, latex_caption)
+              end
+              return result
             else
-              result:insert(1, pandoc.RawBlock("latex", content))
+              local result = pandoc.Blocks({latex_caption, pandoc.RawInline("latex", "\\tabularnewline")})
+              -- if cap_loc is top, insert content on bottom
+              if cap_loc == "top" then
+                result:insert(pandoc.RawBlock("latex", content))        
+              else
+                result:insert(1, pandoc.RawBlock("latex", content))
+              end
+              result:insert(1, pandoc.RawBlock("latex", start))
+              result:insert(1, pandoc.RawBlock("latex", longtable_preamble))
+              result:insert(pandoc.RawBlock("latex", "\\end{longtable}"))
+              result:insert(pandoc.RawBlock("latex", longtable_postamble))
+              return result
             end
-            result:insert(1, pandoc.RawBlock("latex", start))
-            result:insert(1, pandoc.RawBlock("latex", longtable_preamble))
-            result:insert(pandoc.RawBlock("latex", "\\end{longtable}"))
-            result:insert(pandoc.RawBlock("latex", longtable_postamble))
-            return result
           end
-        end
+      end
       end
       return set_raw
     end

src/resources/filters/modules/patterns.lua

@@ -65,6 +65,17 @@ local function match_all_in_table(pattern_table)
   return inner
 end
 
+-- return the pattern, and matched content for the first pattern in the list that matches
+local function match_in_list_of_patterns(raw_tex, list_of_patterns)
+  for _, pattern in ipairs(list_of_patterns) do
+    local matched =  { match_all_in_table(pattern)(raw_tex) }
+    if matched and #matched > 0 then
+      return matched, pattern
+    end
+  end
+  return nil
+end
+
 return {
   attr_identifier = attr_identifier,
   engine_escape = engine_escape,
@@ -93,5 +104,6 @@ return {
   latex_table_star = latex_table_star,
 
   match_all_in_table = match_all_in_table,
+  match_in_list_of_patterns = match_in_list_of_patterns,
   combine_patterns = combine_patterns
 }

src/resources/filters/normalize/flags.lua

@@ -57,11 +57,12 @@ function compute_flags()
       end
 
       if _quarto.format.isRawLatex(el) then
-        local long_table_match = _quarto.modules.patterns.match_all_in_table(_quarto.patterns.latexLongtablePattern)
-        local caption_match = _quarto.modules.patterns.match_all_in_table(_quarto.patterns.latexCaptionPattern)
-        if (long_table_match(el.text) and
-            not caption_match(el.text)) then
-            flags.has_longtable_no_caption_fixup = true
+        local long_table_match, _ = _quarto.modules.patterns.match_in_list_of_patterns(el.text, _quarto.patterns.latexLongtableEnvPatterns)
+        if long_table_match then
+            local caption_match, _= _quarto.modules.patterns.match_in_list_of_patterns(el.text, _quarto.patterns.latexCaptionPatterns)
+            if not caption_match then
+              flags.has_longtable_no_caption_fixup = true
+            end
         end
       end
 

src/resources/filters/quarto-post/latex.lua

@@ -573,12 +573,16 @@ function render_latex_fixups()
   return {{
     RawBlock = function(raw)
       if _quarto.format.isRawLatex(raw) then
-        local long_table_match = _quarto.modules.patterns.match_all_in_table(_quarto.patterns.latexLongtablePattern)
-        local caption_match = _quarto.modules.patterns.match_all_in_table(_quarto.patterns.latexCaptionPattern)
-        if long_table_match(raw.text) and not caption_match(raw.text) then
-          raw.text = raw.text:gsub(
-            _quarto.modules.patterns.combine_patterns(_quarto.patterns.latexLongtablePattern), "\\begin{longtable*}%2\\end{longtable*}", 1)
-          return raw
+        local longtable_match, _ = _quarto.modules.patterns.match_in_list_of_patterns(raw.text, _quarto.patterns.latexLongtableEnvPatterns)
+        if longtable_match then
+          local caption_match = _quarto.modules.patterns.match_in_list_of_patterns(raw.text, _quarto.patterns.latexCaptionPatterns)
+          if not caption_match then
+            -- We need to use the most generic pattern (last of the list) as we want to replace the environment and keep any options 
+            -- (e.g. `\begin{longtable}[c]{ll}` -> \begin{longtable*}[c]{ll} in flextable)
+            local longtable_pattern = _quarto.patterns.latexLongtableEnvPatterns[#_quarto.patterns.latexLongtableEnvPatterns]
+            raw.text = raw.text:gsub(_quarto.modules.patterns.combine_patterns(longtable_pattern), "\\begin{longtable*}%2\\end{longtable*}", 1)
+            return raw
+          end
         end
       end
     end,

src/resources/filters/quarto-pre/parsefiguredivs.lua

@@ -742,11 +742,11 @@ function parse_floatreftargets()
       end
 
       -- finally, if the user passed a \\begin{table} float environment
-      -- we just remove it because we'll re-emit later ourselves
-
-      local b, e, begin_table, table_body, end_table = raw.text:find(patterns.latex_table)
-      if b ~= nil then
-        raw.text = table_body
+      -- we just remove it because we'll re-emit later ourselves  
+      local matched, _ = _quarto.modules.patterns.match_in_list_of_patterns(raw.text, _quarto.patterns.latexTableEnvPatterns)
+      if matched then
+        -- table_body is second matched element.
+        raw.text = matched[2]
       end
 
       return quarto.FloatRefTarget({

src/resources/filters/quarto-pre/table-captions.lua

@@ -29,14 +29,16 @@ function table_captions()
           el = _quarto.ast.walk(el, {
             RawBlock = function(raw)
               if _quarto.format.isRawLatex(raw) then
-                local tabular_match = _quarto.modules.patterns.match_all_in_table(_quarto.patterns.latexTabularPattern)
-                local table_match = _quarto.modules.patterns.match_all_in_table(_quarto.patterns.latexTablePattern)
-                if tabular_match(raw.text) and not table_match(raw.text) then
-                  raw.text = raw.text:gsub(
-                    _quarto.modules.patterns.combine_patterns(_quarto.patterns.latexTabularPattern),
-                    "\\begin{table}\n\\centering\n%1%2%3\n\\end{table}\n",
-                    1)
-                  return raw
+                local tabular_match, tabular_pattern = _quarto.modules.patterns.match_in_list_of_patterns(raw.text, _quarto.patterns.latexTabularEnvPatterns)
+                if tabular_match then 
+                  local table_match, _ = _quarto.modules.patterns.match_in_list_of_patterns(raw.text, _quarto.patterns.latexTableEnvPatterns)
+                  if not table_match then
+                    raw.text = raw.text:gsub(
+                      _quarto.modules.patterns.combine_patterns(tabular_pattern),
+                      "\\begin{table}\n\\centering\n%1%2%3\n\\end{table}\n",
+                      1)
+                    return raw
+                  end
                 end
               end
             end
@@ -169,13 +171,10 @@ function applyTableCaptions(el, tblCaptions, tblLabels)
           raw.text = raw.text:gsub(captionPattern, "%1" .. captionText:gsub("%%", "%%%%") .. "%3", 1)
           idx = idx + 1
         elseif hasRawLatexTable(raw) then
-          for i,pattern in ipairs(_quarto.patterns.latexTablePatterns) do
-            local match_fun = _quarto.modules.patterns.match_all_in_table(pattern)
-            if match_fun(raw.text) then
-              local combined_pattern = _quarto.modules.patterns.combine_patterns(pattern)
+          local matched_env, pattern_env = _quarto.modules.patterns.match_in_list_of_patterns(raw.text, _quarto.patterns.latexAllTableEnvPatterns)
+          if matched_env then
+              local combined_pattern = _quarto.modules.patterns.combine_patterns(pattern_env)
               raw.text = applyLatexTableCaption(raw.text, tblCaptions[idx], tblLabels[idx], combined_pattern)
-              break
-            end
           end
           idx = idx + 1
         elseif hasPagedHtmlTable(raw) then
@@ -200,24 +199,31 @@ end
 
 
 function applyLatexTableCaption(latex, tblCaption, tblLabel, tablePattern)
-  local latexCaptionPattern = _quarto.patterns.latexCaptionPattern
-  local latex_caption_match = _quarto.modules.patterns.match_all_in_table(latexCaptionPattern)
+  local latex_caption_match, _ = _quarto.modules.patterns.match_in_list_of_patterns(latex, _quarto.patterns.latexCaptionPatterns)
   -- insert caption if there is none
-  local beginCaption, caption = latex_caption_match(latex)
-  if not beginCaption then
+  if not latex_caption_match then
     latex = latex:gsub(tablePattern, "%1" .. "\n\\caption{ }\\tabularnewline\n" .. "%2%3", 1)
   end
+  -- caption will be matched
+  latex_caption_match, latex_caption_pattern = _quarto.modules.patterns.match_in_list_of_patterns(latex, _quarto.patterns.latexCaptionPatterns)
   -- apply table caption and label
-  local beginCaption, captionText, endCaption = latex_caption_match(latex)
-  if #tblCaption > 0 then
-    captionText = stringEscape(tblCaption, "latex")
-  end
-  if #tblLabel > 0 then
-    captionText = captionText .. " {#" .. tblLabel .. "}"
+  if not latex_caption_match then
+    -- should never happen as we add the caption command to latex string above
+    -- added to make linter happy too.
+    fatal("Internal Error: \\caption not correctly added in " .. latex)
+  else
+    -- caption text is second element of matched pattern
+    local captionText = latex_caption_match[2]
+    if #tblCaption > 0 then
+      captionText = stringEscape(tblCaption, "latex")
+    end
+    if #tblLabel > 0 then
+      captionText = captionText .. " {#" .. tblLabel .. "}"
+    end
+    assert(captionText)
+    latex = latex:gsub(_quarto.modules.patterns.combine_patterns(latex_caption_pattern), "%1" .. captionText:gsub("%%", "%%%%") .. "%3", 1)
+    return latex
   end
-  assert(captionText)
-  latex = latex:gsub(_quarto.modules.patterns.combine_patterns(latexCaptionPattern), "%1" .. captionText:gsub("%%", "%%%%") .. "%3", 1)
-  return latex
 end