Feature text enrichment (#29)

* inline formatting, better inlining handling overall * first iteration of refactoring * only codeblocks not working * refactoring somewhat complete for now --------- Co-authored-by: varphi-online <[email protected]>
TimoKats · Sep 16, 2024 · 07e0235 · 07e0235
1 parent b9472dd
commit 07e0235
Show file tree

Hide file tree

Showing 4 changed files with 152 additions and 85 deletions.
diff --git a/lib/converter.go b/lib/converter.go
@@ -0,0 +1,99 @@
+package lib
+
+import (
+  "strings"
+  "regexp"
+)
+
+var markdownUlListActive bool
+var markdownOlIndex int64
+var codeBlockActive = false
+var codeBlockAggregate = ""
+
+func ConvertReferenceLink(text string, markdownLinks *regexp.Regexp) string {
+  return "<p>" + markdownLinks.ReplaceAllString(text, "<a href='$2'>$1</a>") + "</p>"
+}
+
+func ConvertEnclosure(text string, markdownLinks *regexp.Regexp) string {
+  url := markdownLinks.ReplaceAllString(text, "$2")
+  size, fileSizeErr := FileSizeUrl(url)
+  if fileSizeErr != nil {
+    Error.Println(fileSizeErr)
+    return ""
+  }
+  return "<enclosure " + markdownLinks.ReplaceAllString(text, "url='$2' type='$1' length='") + size + "' />"
+}
+
+func ConvertUnorderedlList(text string, markdownUnorderedLists *regexp.Regexp) string {
+  if markdownUnorderedLists.MatchString(text) {
+    if !markdownUlListActive {
+      markdownUlListActive = true
+      return "<ul><li>" + text[getLeadingWhitespace(text):] + "</li>"
+    }
+    return "<li>" + text[getLeadingWhitespace(text):] + "</li>"
+  }
+  markdownUlListActive = false
+  return "</ul><p>" + text + "</p>"
+}
+
+
+func ConvertTextEnrichment(text string) string {
+  // Strikethrough
+  text = inlineRewrap(text, regexp.MustCompile(`(~~)([^~~]+)(~~)`), "<s>", "</s>")
+  // Subscript
+  text = inlineRewrap(text, regexp.MustCompile(`(~)([^~]+)(~)`), "<sub>", "</sub>")
+  // Superscript
+  text = inlineRewrap(text, regexp.MustCompile(`(\^)([^\^]+)(\^)`), "<sup>", "</sup>")
+  // Strong Emphasis
+  text = inlineRewrap(text, regexp.MustCompile(`(\*\*\*|___)([^\*^_]+)(\*\*\*|___)`), "<b><i>", "</b></i>")
+  // Bold
+  text = inlineRewrap(text, regexp.MustCompile(`(\*\*|__)([^\*^_]+)(\*\*|__)`), "<b>", "</b>")
+  // Italic
+  text = inlineRewrap(text, regexp.MustCompile(`(\*|_)([^\*^_]+)(\*|_)`), "<i>", "</i>")
+  // Inline Codeblock
+  text = inlineRewrap(text, regexp.MustCompile(`(\x60)([^\x60]+)(\x60)`), "<code>", "</code>")
+  return text
+}
+
+func ConvertLink(text string, markdownLinks *regexp.Regexp) string {
+  if strings.Contains(text, "audio/mpeg") {
+    return ConvertEnclosure(text, markdownLinks)
+  } else {
+    return ConvertReferenceLink(text, markdownLinks)
+  }
+}
+
+func ConvertCodeblock(text string, fencedCodeBlock *regexp.Regexp) string {
+  if !codeBlockActive {
+    codeBlockActive = true
+    codeBlockAggregate = "<sup>" + text[3:] + "</sup><br>"
+    return "" + "<pre style=\"word-wrap: break-word;\"><code>"
+  } else if fencedCodeBlock.MatchString(text) {
+    out := codeBlockAggregate
+    codeBlockAggregate, codeBlockActive = "", false
+    return out + "</code></pre>"
+  }
+  if codeBlockAggregate != "" {
+    codeBlockAggregate += "<br>"
+  }
+  codeBlockAggregate += text
+  return ""
+}
+
+func ConvertOrderedlList(text string) string {
+  if markdownOlIndex == 0 {
+    markdownOlIndex = 1
+    return "<ol><li>" + text[getLeadingWhitespace(text):] + "</li>"
+  }
+  return "<li>" + text[getLeadingWhitespace(text):] + "</li>"
+}
+
+func ConvertOrderedLists(text string, markdownOrderedLists *regexp.Regexp) string {
+  if markdownOrderedLists.MatchString(text) {
+    entryText := markdownOrderedLists.FindStringSubmatch(text)[4]
+    return ConvertOrderedlList(entryText)
+  }
+  markdownOlIndex = 0
+  return "</ol>" + ConvertMarkdownToRSS(text)
+}
+
diff --git a/lib/markdown.go b/lib/markdown.go
@@ -1,19 +1,14 @@
 package lib
 
 import (
+  "os"
   "bufio"
   "errors"
-  "os"
   "regexp"
-  "strconv"
   "strings"
   "unicode"
 )
 
-var markdownUlListActive bool
-var markdownOlIndex int64
-var codeBlockAggregate, codeBlockOpen = "", false
-
 func checkMarkdownTitle(text string) bool {
   if len(text) > 0 {
     return string(text[0]) == string("#")
@@ -30,88 +25,37 @@ func getLeadingWhitespace(text string) int {
   return 0
 }
 
-func convertMarkdownLink(text string, markdownLinks *regexp.Regexp) string {
-  return "<p>" + markdownLinks.ReplaceAllString(text, "<a href='$2'>$1</a>") + "</p>"
-}
-
-func convertMarkdownEnclosure(text string, markdownLinks *regexp.Regexp) string {
-  url := markdownLinks.ReplaceAllString(text, "$2")
-  size, fileSizeErr := FileSizeUrl(url)
-  if fileSizeErr != nil {
-    Error.Println(fileSizeErr)
-    return ""
-  }
-  return "<enclosure " + markdownLinks.ReplaceAllString(text, "url='$2' type='$1' length='") + size + "' />"
-}
-
-func convertMarkdownUlList(text string) string {
-  if !markdownUlListActive {
-    markdownUlListActive = true
-    return "<ul><li>" + text[getLeadingWhitespace(text):] + "</li>"
-  }
-  return "<li>" + text[getLeadingWhitespace(text):] + "</li>"
-}
-
-func convertMarkdownOlList(text string, index int64) string {
-  if markdownOlIndex == 0 {
-    markdownOlIndex = 1
-    return "<ol><li>" + text[getLeadingWhitespace(text):] + "</li>"
-  }
-  return "<li>" + text[getLeadingWhitespace(text):] + "</li>"
-}
-
 func ConvertMarkdownToRSS(text string) string {
   markdownLinks := regexp.MustCompile(`\[(.*)\]\((.*)\)`)
   markdownUnorderedLists := regexp.MustCompile(`^(\s*)(-|\*|\+)[\s](.*)`)
   markdownOrderedLists := regexp.MustCompile(`^(\s*)(-?\d+)(\.\s+)(.*)$`)
   fencedCodeBlock := regexp.MustCompile("^\x60\x60\x60")
-  inlineCodeBlock := regexp.MustCompile(`([\x60]+)([^\x60]+)([\x60]+)`)
-
+  text = ConvertTextEnrichment(text)
   switch {
-    case codeBlockOpen && !fencedCodeBlock.MatchString(text):
-      if codeBlockAggregate != "" {
-        codeBlockAggregate += "<br>"
-      }
-      codeBlockAggregate += text
-      return ""
+    // links 
     case markdownLinks.MatchString(text):
-      if strings.Contains(text, "audio/mpeg") {
-        return convertMarkdownEnclosure(text, markdownLinks)
-      } else {
-        return convertMarkdownLink(text, markdownLinks)
-      }
-    case markdownUnorderedLists.MatchString(text):
-      return convertMarkdownUlList(text)
-    case markdownUlListActive:
-      markdownUlListActive = false
-      return "</ul><p>" + text + "</p>"
-    case markdownOrderedLists.MatchString(text):
-      entryIndex, entryErr := strconv.ParseInt(markdownOrderedLists.FindStringSubmatch(text)[2], 10, 64)
-      entryText := markdownOrderedLists.FindStringSubmatch(text)[4]
-      if entryErr != nil {
-        return "<p>" + text + "</p>"
-      }
-      return convertMarkdownOlList(entryText, entryIndex)
-    case markdownOlIndex != 0:
-      markdownOlIndex = 0
-      return "</ol>" + ConvertMarkdownToRSS(text)
-    case fencedCodeBlock.MatchString(text):
-      if !codeBlockOpen {
-        codeBlockOpen = true
-        codeBlockAggregate = "<sup>" + text[3:] + "</sup><br>"
-        return "" + "<pre style=\"word-wrap: break-word;\"><code>"
-      } else {
-        out := codeBlockAggregate
-        codeBlockAggregate, codeBlockOpen = "", false
-        return out + "</code></pre>"
-      }
-    case inlineCodeBlock.Match([]byte(text)):
-      out := inlineCodeBlock.ReplaceAllFunc([]byte(text), func(b []byte) []byte {
-        return []byte("<code>" + inlineCodeBlock.FindStringSubmatch(string(b))[2] + "</code>")
-      })
-      return string(out)
-    default:
-      return "<p>" + text + "</p>"
+      return ConvertLink(text, markdownLinks)
+    // lists
+    case markdownUnorderedLists.MatchString(text) || markdownUlListActive:
+      return ConvertUnorderedlList(text, markdownUnorderedLists)
+    case markdownOrderedLists.MatchString(text) || markdownOlIndex != 0:
+      return ConvertOrderedLists(text, markdownOrderedLists)
+    //code blocks
+    case fencedCodeBlock.MatchString(text) || codeBlockActive:
+      return ConvertCodeblock(text, fencedCodeBlock)
+  default:
+    return "<p>" + text + "</p>"
+  }
+}
+
+func inlineRewrap(text string, pattern *regexp.Regexp, prefix string, postfix string) string {
+  if pattern.Match([]byte(text)) {
+    out := pattern.ReplaceAllFunc([]byte(text), func(b []byte) []byte {
+      return []byte(prefix + pattern.FindStringSubmatch(string(b))[2] + postfix)
+    })
+    return string(out)
+  } else {
+    return text
   }
 }
 
@@ -142,7 +86,7 @@ func ReadMarkdown(config Config, articles []Article) []Article {
     for scanner.Scan() {
       if checkMarkdownTitle(scanner.Text()) && len(articles[index].Title) == 0 {
         articles[index].Title = scanner.Text()[2:len(scanner.Text())]
-      } else if len(scanner.Text()) > 0 || codeBlockOpen {
+      } else if len(scanner.Text()) > 0 || codeBlockActive {
         articles[index].Description += ConvertMarkdownToRSS(scanner.Text())
       }
     }

diff --git a/lib/types.go b/lib/types.go
@@ -1,9 +1,26 @@
 package lib
 
 import (
+  "regexp"
   "time"
 )
 
+type Markdown struct {
+  Content []Line
+}
+
+type Line struct {
+  Link *regexp.Regexp
+  UnorderedList *regexp.Regexp
+  OrderedList *regexp.Regexp
+  CodeBlock *regexp.Regexp
+
+  // optional fields
+  CodeBlockOpen bool
+  UnorderedListActive bool
+  OrderedListActive bool
+}
+
 type Article struct {
   Id int
   Title string

diff --git a/test/another-article.md b/test/another-article.md
@@ -13,7 +13,14 @@ Timo
 
 And here goes some text...
 
-[link](https://timokats.xyz)
+**bold** text, __however you like it__.
+*italic* too? _Of course_!
+Sometimes ***strong emphasis*** is needed to get across a ___point___.
+~~You can always striketrough a bad idea~~,
+Make something unique with a ~subscript~,
+Or power up with a ^superscript^!
+
+A [link](https://timokats.xyz)
 
 And this is a list:
 - hello
@@ -36,6 +43,8 @@ fencedCodeBlock := regexp.MustCompile("^```")
 	  // Show off your tabwidth in style!
 ```
 
+Can I write text in between the code blocks?
+
 ```C++
 int main() {
     int n, t1 = 0, t2 = 1, nextTerm = 0;
@@ -68,5 +77,3 @@ int main() {
 This feature also works `inline` as well!
 
 And back to text again
-
-