add asciidoc style for vale to be able to parse raw content to find markup syntax as well as content

Author: rosieyohannan

.vale.ini

@@ -4,7 +4,7 @@ Vocab = Docs
 
 MinAlertLevel = suggestion
 
-Packages = Readability
+Packages = Readability, AsciiDoc
 
 [asciidoctor]
 # attribute = value
@@ -18,4 +18,4 @@ experimental = YES
 attribute-missing = drop
 
 [*.adoc]
-BasedOnStyles = circleci-docs, Vale, Readability, Openly
+BasedOnStyles = circleci-docs, Vale, Readability, Openly, AsciiDoc

styles/AsciiDoc/ClosedAttributeBlocks.yml

@@ -0,0 +1,7 @@
+---
+extends: existence
+scope: raw
+level: error
+message: "Attribute block is not closed."
+raw:
+  - '(?<!.)(\[(?!{)[^\]\n]+)\n'

styles/AsciiDoc/ClosedIdQuotes.yml

@@ -0,0 +1,7 @@
+---
+extends: existence
+scope: raw
+level: error
+message: "Quoted ID value is not closed."
+raw:
+  - '(?<!.)\[id=(["\x27]).*(?<!\1)\]'

styles/AsciiDoc/ImageContainsAltText.yml

@@ -0,0 +1,8 @@
+---
+extends: existence
+scope: raw
+level: warning
+link: https://redhat-documentation.github.io/supplementary-style-guide/#cloud-services-images
+message: "Image is missing accessibility alt tags."
+raw:
+  - 'image::?.*\[\]'

styles/AsciiDoc/LinkContainsLinkText.yml

@@ -0,0 +1,10 @@
+---
+extends: existence
+scope: raw
+level: warning
+ignorecase: true
+nonword: true
+link: https://redhat-documentation.github.io/supplementary-style-guide/#cloud-services-links-hypertext
+message: "Link is missing descriptive link text for accessibility."
+raw:
+  - 'link:.*(\[\]|\[click here\]|\[here\]|\[this\])'

styles/AsciiDoc/MatchingDotCallouts.yml

@@ -0,0 +1,47 @@
+---
+extends: script
+message: "Corresponding callout not found."
+level: warning
+link: https://docs.asciidoctor.org/asciidoc/latest/verbatim/callouts/
+scope: raw
+script: |
+  text := import("text")
+  matches := []
+
+  // clean out multi-line comments
+  scope = text.re_replace("(?s) *(\n////.*?////\n)", scope, "")
+  //add a newline, it might be missing
+  scope += "\n"
+
+  num_codeblock_callouts := 0
+  num_callouts := 0
+  codeblock_callout_regex := ".+<(\\.)>"
+  callout_regex := "^<(\\.)>"
+
+  for line in text.split(scope, "\n") {
+    // trim trailing whitespace
+    line = text.trim_space(line)
+    if text.re_match(codeblock_callout_regex, line) {
+      //restart for new listingblock
+      num_callouts = 0
+      //account for lines with multiple callouts
+      num_callouts_in_line := text.count(line, "<.>")
+      if num_callouts_in_line > 1 {
+        num_codeblock_callouts = num_codeblock_callouts + num_callouts_in_line
+      } else {
+        num_codeblock_callouts++
+      }
+    }
+
+    if text.re_match(callout_regex, line) {
+      num_callouts++
+      if num_callouts > num_codeblock_callouts {
+        start := text.index(scope, line)
+        matches = append(matches, {begin: start, end: start + len(line)})
+      }
+      if num_callouts == num_codeblock_callouts {
+        num_callouts = 0
+        num_codeblock_callouts = 0
+      }
+    }
+  }

styles/AsciiDoc/MatchingNumberedCallouts.yml

@@ -0,0 +1,62 @@
+---
+extends: script
+message: "Corresponding callout not found."
+level: warning
+link: https://docs.asciidoctor.org/asciidoc/latest/verbatim/callouts/
+scope: raw
+script: |
+  text := import("text")
+  matches := []
+
+  // clean out multi-line comments
+  scope = text.re_replace("(?s) *(\n////.*?////\n)", scope, "")
+  //add a newline, it might be missing
+  scope += "\n"
+
+  codeblock_callout_regex := ".+(<\\d+>)+"
+  callout_regex := "^<(\\d+)>"
+  codeblock_callouts := []
+  inside := false
+  found := false
+  num_callouts := 0
+
+  num_lines := len(text.split(scope, "\n"))
+
+  for line in text.split(scope, "\n") {
+    // trim trailing whitespace
+    line = text.trim_space(line)
+    if text.re_match(codeblock_callout_regex, line) {
+      inside = true
+      //account for lines with multiple callouts
+      for i := 1; i < num_lines; i++ {
+        //text.contains must be str, not regex
+        str := "<" + i + ">"
+        if text.contains(line, str) {
+          codeblock_callouts = append(codeblock_callouts, i)
+        }
+      }
+    } else if text.re_match(callout_regex, line) {
+      inside = false
+      found = false
+      num_callouts = 1
+      for i in codeblock_callouts {
+        //cast int > string
+        j := text.itoa(i)
+        str := "<" + j + ">"
+        if text.contains(line, str) {
+          //setting found allows us to loop through the list of possible matches
+          found = true
+        }
+        num_callouts++
+      }
+      if !found {
+        start := text.index(scope, line)
+        matches = append(matches, {begin: start, end: start + len(line)})
+      }
+    } else if codeblock_callouts && inside == false {
+      //cycled through num_callouts, reset for next codeblock
+      if num_callouts == len(codeblock_callouts) {
+        codeblock_callouts = []
+      }
+    }
+  }

styles/AsciiDoc/SequentialNumberedCallouts.yml

@@ -0,0 +1,59 @@
+---
+extends: script
+message: "Numbered callout does not follow sequentially."
+level: warning
+link: https://docs.asciidoctor.org/asciidoc/latest/verbatim/callouts/
+scope: raw
+script: |
+  text := import("text")
+  matches := []
+
+  // clean out multi-line comments
+  scope = text.re_replace("(?s) *(\n////.*?////\n)", scope, "")
+  //add a newline, it might be missing
+  scope += "\n"
+
+  prev_num := 0
+  callout_regex := "^<(\\d+)>"
+  listingblock_delim_regex := "^-{4,}$"
+  if_regex := "^ifdef::|ifndef::"
+  endif_regex := "^endif::\\[\\]"
+  inside_if := false
+
+  for line in text.split(scope, "\n") {
+    // trim trailing whitespace
+    line = text.trim_space(line)
+
+    // check if we're entering a conditional block
+    if text.re_match(if_regex, line) {
+        inside_if = true
+    } else if text.re_match(endif_regex, line) {
+        inside_if = false
+    }
+
+    //reset count if we hit a listing block delimiter
+    if text.re_match(listingblock_delim_regex, line) {
+      prev_num = 0
+    }
+
+    //only count callouts where there are no ifdefs
+    if !inside_if {
+      if text.re_match(callout_regex, line) {
+        callout := text.re_find("<(\\d+)>", line)
+        for key, value in callout {
+          //trim angle brackets from string
+          trimmed := callout[key][0]["text"]
+          trimmed = text.trim_prefix(trimmed, "<")
+          trimmed = text.trim_suffix(trimmed, ">")
+          //cast string > int
+          num := text.atoi(trimmed)
+          //start counting
+          if num != prev_num+1 {
+            start := text.index(scope, line)
+            matches = append(matches, {begin: start, end: start + len(line)})
+          }
+          prev_num = num
+        }
+      }
+    }
+  }

styles/AsciiDoc/UnsetAttributes.yml

@@ -0,0 +1,49 @@
+---
+extends: script
+level: suggestion
+message: "Set attribute directive does not have a corresponding unset attribute directive."
+link: https://docs.asciidoctor.org/asciidoc/latest/attributes/unset-attributes/#unset-a-document-attribute-in-the-body
+scope: raw
+script: |
+  text := import("text")
+  matches := []
+
+  // trim extra whitespace
+  scope = text.trim_space(scope)
+  // add a newline, it might be missing
+  scope += "\n"
+  // clean out multi-line comments
+  scope = text.re_replace("(?s) *(\n////.*?////\n)", scope, "")
+
+  attr_regex := "^:[\\w-_]+:.*$"
+  context_mod_docs_regex := "^:context|_content-type|_mod-docs-content-type:.*$"
+  attr_name_regex := ":[\\w-_]+:"
+  attr_name := ""
+  unset_attr_pref := ""
+  unset_attr_suff := ""
+
+  for line in text.split(scope, "\n") {
+    if text.re_match(attr_regex, line) {
+      if !text.re_match(context_mod_docs_regex, line) {
+        start := text.index(scope, line)
+        matches = append(matches, {begin: start, end: start + len(line)})
+        // re_find returns an array holding all matches
+        attr_name = ((text.re_find(attr_name_regex, line))[0][0])["text"]
+        unset_attr_pref = `^:!` + text.trim_prefix(attr_name, `:`)
+        unset_attr_suff = `^` + text.trim_suffix(attr_name, `:`) + `!:`
+        // loop through lines for every attr found
+        for line in text.split(scope, "\n") {
+          if text.re_match(unset_attr_pref, line) {
+            if len(matches) > 0 {
+              // remove the most recently added match
+              matches = matches[:len(matches)-1]
+            } else if text.re_match(unset_attr_suff, line) {
+              if len(matches) > 0 {
+                matches = matches[:len(matches)-1]
+              }
+            }
+          }
+        }
+      }
+    }
+  }

styles/AsciiDoc/ValidAdmonitionBlocks.yml

@@ -0,0 +1,30 @@
+---
+extends: script
+level: error
+message: "Unterminated admonition block found in file."
+link: https://docs.asciidoctor.org/asciidoc/latest/blocks/admonitions/
+scope: raw
+script: |
+  text := import("text")
+  matches := []
+
+  // clean out multi-line comments
+  scope = text.re_replace("(?s) *(\n////.*?////\n)", scope, "")
+  //add a newline, it might be missing
+  scope += "\n"
+
+  admon_delim_regex := "^={4}$"
+  count := 0
+
+  for line in text.split(scope, "\n") {
+    // trim trailing whitespace
+    line = text.trim_space(line)
+    if text.re_match(admon_delim_regex, line) {
+      count += 1
+      start := text.index(scope, line)
+      matches = append(matches, {begin: start, end: start + len(line)})
+    } else if count > 1 {
+      count = 0 // admonition block is closed, reset the count
+      matches = []
+    }
+  }