Enable line-breaking blockquotes (prefaced by |)

TODO: No indentation on first line

data/ms_odt.xsl

@@ -78,7 +78,7 @@
  </xsl:template>
 
  <xsl:template match="ser:p" mode="quote">
-     <text:p text:style-name="Quotations"><text:span text:style-name="T2"><xsl:apply-templates select="*|text()"/></text:span></text:p>
+     <text:p text:style-name="Quotations"><xsl:apply-templates select="*|text()"/></text:p>
  </xsl:template>
 
  <xsl:template match="ser:quote" mode="body">

src/format.c

@@ -45,6 +45,7 @@ typedef enum {
     TOK_STAR,
     TOK_REF,
     TOK_URL,
+    TOK_BREAK,
     /*
     TOK_DASH,
     TOK_OPEN_DOUBLE_QUOTE,
@@ -115,6 +116,11 @@ nextToken(Tokenizer tokenizer) {
         result.toktype = TOK_STAR;
         result.toktext = NULL;
         return result;
+    } else if (ch == '\n') {
+        utf8Advance(tokenizer);
+        result.toktype = TOK_BREAK;
+        result.toktext = NULL;
+        return result;
     } else if (greekChar(ch)) {
         while ((ch != 0) &&
                (greekChar(ch) || (ch == ' ') || (ch == ',') || (ch == '.'))) {
@@ -158,7 +164,7 @@ nextToken(Tokenizer tokenizer) {
         result.toktext = strndup(tokenizer->txt + startIndex, endIndex - startIndex);
         return result;
     } else if (latinChar(ch)) {
-        while ((ch != 0) && latinChar(ch) && (ch != '*')) {
+        while ((ch != 0) && latinChar(ch) && (ch != '*') && (ch != '\n')) {
             utf8Advance(tokenizer);
             ch = utf8CharAt(tokenizer);
             if (ch == '^') {
@@ -206,6 +212,10 @@ int formatText(const char* txt, FormatElement** dst, CitationRecordQueue* citati
                 REINIT_QUEUE(formatElementQ);
                 em = 1;
             }
+        } else if (tok.toktype == TOK_BREAK) {
+            FormatElement elt = { .elementType = FORMAT_BR, .elementContentLength = 0,
+                                  .elementContent = { .textContent = NULL } };
+            APPEND_QUEUE(FormatElementQueue, formatElementQ, elt);
         } else {
             FormatElementType t;
             FormatElement elt = { .elementContent = { .textContent = tok.toktext } } ;

src/format.h

@@ -34,6 +34,7 @@
 typedef enum {
     FORMAT_TEXT,
     FORMAT_EM,
+    FORMAT_BR,
     FORMAT_STRONG,
     FORMAT_CITATION,
     FORMAT_GREEK,

src/sermon.h

@@ -38,7 +38,8 @@ typedef struct {
 
 typedef enum {
     PARA_DEFAULT,
-    PARA_BLOCKQUOTE
+    PARA_BLOCKQUOTE,
+    PARA_BLOCKPRESERVE
 } SermonParagraphType;
 
 typedef struct {

src/sermon_lexer.l

@@ -24,14 +24,23 @@ ID                           [A-Za-z_][A-Za-z_0-9]*
 <REFERENCEINTRO>ref          { yylloc.first_column = yylloc.last_column + 1; yylloc.last_column = yylloc.first_column + 2; return KW_REF; }
 <REFERENCEINTRO>:            { BEGIN(REFERENCENAME); yylloc.first_column = ++yylloc.last_column; return ':'; }
 <REFERENCENAME>{ID}          { yylloc.first_column = yylloc.last_column + 1; yylloc.last_column = yylloc.first_column + strlen(yytext) + 1; yylval.sval = strdup(yytext); return ID; }
-<REFERENCENAME>:            { BEGIN(REFERENCE); yylloc.first_column = ++yylloc.last_column; return ':'; }
+<REFERENCENAME>:             { BEGIN(REFERENCE); yylloc.first_column = ++yylloc.last_column; return ':'; }
-<REFERENCE>[^}]*            { yylloc.first_column = yylloc.last_column + 1; yylloc.last_column = yylloc.first_column + strlen(yytext) + 1; yylval.sval = strdup(yytext); return REFTEXT; }
+<REFERENCE>[^}]*             { yylloc.first_column = yylloc.last_column + 1; yylloc.last_column = yylloc.first_column + strlen(yytext) + 1; yylval.sval = strdup(yytext); return REFTEXT; }
-<REFERENCE>[}]              { BEGIN(INITIAL); yylloc.first_column = ++yylloc.last_column; return '}'; }
+<REFERENCE>[}]               { BEGIN(INITIAL); yylloc.first_column = ++yylloc.last_column; return '}'; }
-^[^[{>\n].*                   { yylloc.first_column = ++yylloc.last_column; yylval.sval = strdup(yytext); LEXPRINT("LINE: %s\n", yytext); return LINE; }
+^[^[{>|\n].*                 { yylloc.first_column = ++yylloc.last_column; yylval.sval = strdup(yytext); LEXPRINT("LINE: %s\n", yytext); return LINE; }
-^>{WHITESPACE}*\n             { yylloc.first_column = yylloc.last_column = 0; yylloc.first_line++; yylloc.last_line++; return '\n'; }
+^[>|]{WHITESPACE}*\n         { yylloc.first_column = yylloc.last_column = 0; yylloc.first_line++; yylloc.last_line++; return '\n'; }
-^>{WHITESPACE}*               { BEGIN(BLOCK); yylloc.first_column = yylloc.last_column = 1; return '>'; }
+^[>|]{WHITESPACE}*           {
-<BLOCK>..*                   { BEGIN(INITIAL); yylloc.first_column = yylloc.last_column + 1; yylloc.last_column = yylloc.first_column + strlen(yytext) - 1; yylval.sval = strdup(yytext); return LINE; }
+                                 BEGIN(BLOCK);
+                                 yylloc.first_column = yylloc.last_column = 1;
+                                 return yytext[0];
+                             }
+<BLOCK>..*                   {
+                                 BEGIN(INITIAL);
+                                 yylloc.first_column = yylloc.last_column + 1;
+                                 yylloc.last_column = yylloc.first_column + strlen(yytext) - 1;
+                                 yylval.sval = strdup(yytext); return LINE;
+                             }
 <INITIAL>\n                  { yylloc.first_column = yylloc.last_column = 0; yylloc.first_line++; yylloc.last_line++; return '\n'; }
-<<EOF>>                     { return 0; }
+<<EOF>>                      { return 0; }
 %%
 

src/sermon_parser.y

@@ -37,7 +37,13 @@ char* lineQueueToString(LineQueue* lq) {
     FOREACH_QUEUE(LineQueue, *lq, ptr) {
         strncat(dest + idx, ptr->data, paraLength - idx);
         idx += strlen(ptr->data);
-        if (ptr->next != NULL) dest[idx++] = ' ';
+        if (ptr->next != NULL) {
+            if (dest[idx-1] != '\n')
+                dest[idx++] = ' ';
+        } else {
+            if (dest[idx-1] == '\n')
+               idx--;
+        }
         free(ptr->data);
     }
     FOREACH_QUEUE_END
@@ -45,6 +51,14 @@ char* lineQueueToString(LineQueue* lq) {
     return dest;
 }
 
+char *lineWithBreak(const char *txt) {
+    size_t l = strlen(txt);
+    char* s = (char*)malloc(l+2);
+    strncpy(s, txt, l + 2);
+    s[l] = '\n';
+    return s;
+}
+
 void yyerror(Sermon*, const char*);
 
 %}
@@ -126,7 +140,7 @@ block:
          PARSEPRINT("Parsed paragraph:\n%s\n\n", p.paraText);
          free(paraText);
      }
-   | blockquote {
+   | blockquote_or_preserve {
          SermonParagraph p = { .paraType = PARA_BLOCKQUOTE };
          char* paraText = lineQueueToString(&lineQ);
          FormatElement* paraContent = NULL;
@@ -142,10 +156,17 @@ para:
     para LINE '\n' { APPEND_QUEUE(LineQueue, lineQ, $2); }
   | LINE '\n'      { DESTROY_QUEUE(LineQueue, lineQ); APPEND_QUEUE(LineQueue, lineQ, $1); }
   ;
+blockquote_or_preserve:
+          blockquote
+        | blockpreserve
+        ;
 blockquote:
           blockquote '>' LINE '\n' { APPEND_QUEUE(LineQueue, lineQ, $3); }
         | '>' LINE '\n' { DESTROY_QUEUE(LineQueue, lineQ); APPEND_QUEUE(LineQueue, lineQ, $2); }
         ;
+blockpreserve:
+          blockpreserve '|' LINE '\n' { char* s = lineWithBreak($3); APPEND_QUEUE(LineQueue, lineQ, s); free($3); }
+        | '|' LINE '\n' { char* s = lineWithBreak($2); DESTROY_QUEUE(LineQueue, lineQ); APPEND_QUEUE(LineQueue, lineQ, s); free($2); }
 references:
           references reference break
         | /* empty */

src/xml.c

@@ -83,6 +83,8 @@ formatElementsToXML(
             xmlNodePtr em = xmlNewNode(sermon_ns, "em");
             formatElementsToXML(sermon_ns, em, a[i].elementContent.nestedContent, a[i].elementContentLength, numReferences, sermonReferencesPtr);
             xmlAddChild(parentElement, em);
+        } else if (a[i].elementType == FORMAT_BR) {
+            xmlAddChild(parentElement, xmlNewNode(sermon_ns, "br"));
         } else if (a[i].elementType == FORMAT_TEXT) {
             xmlAddChild(parentElement, xmlNewText(a[i].elementContent.textContent));
         } else if (a[i].elementType == FORMAT_GREEK) {