Make links live again

src/format.c

@@ -44,6 +44,7 @@ typedef enum {
     TOK_UNICODE,
     TOK_STAR,
     TOK_REF,
+    TOK_URL,
     /*
     TOK_DASH,
     TOK_OPEN_DOUBLE_QUOTE,
@@ -84,6 +85,21 @@ latinChar(uint32_t ch) {
     return (ch <= 0xff) || extendedPunctuation(ch);
 }
 
+inline int
+httpAt(Tokenizer tokenizer) {
+    return ((tolower(tokenizer->txt[tokenizer->byteIndex]) == 'h') &&
+            (tolower(tokenizer->txt[tokenizer->byteIndex + 1]) == 't') &&
+            (tolower(tokenizer->txt[tokenizer->byteIndex + 2]) == 't') &&
+            (tolower(tokenizer->txt[tokenizer->byteIndex + 3]) == 'p') &&
+            ((tokenizer->txt[tokenizer->byteIndex + 4] == ':') &&
+             (tokenizer->txt[tokenizer->byteIndex + 5] == '/') &&
+             (tokenizer->txt[tokenizer->byteIndex + 6] == '/')) ||
+            ((tolower(tokenizer->txt[tokenizer->byteIndex + 4]) == 's') &&
+             (tokenizer->txt[tokenizer->byteIndex + 5] == ':') &&
+             (tokenizer->txt[tokenizer->byteIndex + 6] == '/') &&
+             (tokenizer->txt[tokenizer->byteIndex + 7] == '/')));
+}
+
 static Token
 nextToken(Tokenizer tokenizer) {
     int startIndex = tokenizer->byteIndex;
@@ -125,12 +141,30 @@ nextToken(Tokenizer tokenizer) {
         result.toktype = TOK_REF;
         result.toktext = strndup(tokenizer->txt + idStart, idEnd - idStart);
         return result;
+    } else if (httpAt(tokenizer)) {
+        int endIndex = 0;
+        while ((ch != 0) && (ch != ' ') && (ch != '\r') && (ch != '\n')) {
+            utf8Advance(tokenizer);
+            ch = utf8CharAt(tokenizer);
+        }
+        if (tokenizer->txt[tokenizer->byteIndex - 1] == '.') {
+            /* heuristic: url doesn't end in . */
+            endIndex = --tokenizer->byteIndex;
+        } else {
+            endIndex = tokenizer->byteIndex;
+        }
+
+        result.toktype = TOK_URL;
+        result.toktext = strndup(tokenizer->txt + startIndex, endIndex - startIndex);
+        return result;
     } else if (latinChar(ch)) {
         while ((ch != 0) && latinChar(ch) && (ch != '*')) {
             utf8Advance(tokenizer);
             ch = utf8CharAt(tokenizer);
             if (ch == '^') {
                 if (tokenizer->txt[tokenizer->byteIndex + 1] == '{') break;
+            } else if (httpAt(tokenizer)) {
+                break;
             }
         }
         result.toktype = TOK_TEXT;
@@ -181,6 +215,8 @@ int formatText(const char* txt, FormatElement** dst, CitationRecordQueue* citati
                 t = FORMAT_GREEK;
             } else if (tok.toktype == TOK_UNICODE) {
                 t = FORMAT_UNICODE;
+            } else if (tok.toktype == TOK_URL) {
+                t = FORMAT_URL;
             } else if (tok.toktype == TOK_REF) {
                 t = FORMAT_CITATION;
                 if (citationQPtr && !lookupCitation(*citationQPtr, tok.toktext)) {

src/format.h

@@ -37,7 +37,8 @@ typedef enum {
     FORMAT_STRONG,
     FORMAT_CITATION,
     FORMAT_GREEK,
-    FORMAT_UNICODE
+    FORMAT_UNICODE,
+    FORMAT_URL
 } FormatElementType;
 
 typedef struct FormatElement FormatElement;

src/xml.c

@@ -93,6 +93,11 @@ formatElementsToXML(
             xmlNodePtr unicode = xmlNewNode(sermon_ns, "unicode");
             xmlAddChild(unicode, xmlNewText(a[i].elementContent.textContent));
             xmlAddChild(parentElement, unicode);
+        } else if (a[i].elementType == FORMAT_URL) {
+            xmlNodePtr link = xmlNewNode(sermon_ns, "link");
+            xmlSetProp(link, "href", a[i].elementContent.textContent);
+            xmlAddChild(link, xmlNewText(a[i].elementContent.textContent));
+            xmlAddChild(parentElement, link);
         } else if (a[i].elementType == FORMAT_CITATION) {
             xmlNodePtr cite = xmlNewNode(sermon_ns, "cite");
             int num = findReferenceNumber(numReferences, sermonReferencesPtr, a[i].elementContent.textContent);