Changeset 5992e0e in mainline

Timestamp:

2011-08-19T18:20:04Z (13 years ago)

Author:

Martin Sucha <sucha14@…>

Branches:

lfn, master, serial, ticket/834-toolchain-update, topic/msim-upgrade, topic/simplify-dev-export

Children:

e14a103

Parents:

0662451

Message:

Fix bdsh tokenizer

File:

• uspace/app/bdsh/tok.c (modified) (9 diffs)

uspace/app/bdsh/tok.c

@@ -92 +92 @@
 {
         int rc;
-        wchar_t cur_char;
+        wchar_t next_char;
         
         /* Read the input line char by char and append tokens */
-        while ((cur_char = tok_get_char(tok)) != 0) {
-                if (cur_char == ' ') {
+        while ((next_char = tok_look_char(tok)) != 0) {
+                if (next_char == ' ') {
                         /* Push the token if there is any.
                          * There may not be any pending char for a token in case
@@ -108 +108 @@
                         }
                         tok_start_token(tok, TOKTYPE_SPACE);
-                        /* Eat all spaces */
+                        /* Eat all the spaces */
                         while (tok_look_char(tok) == ' ') {
                                 tok_push_char(tok, tok_get_char(tok));
@@ -115 +115 @@
                         
                 }
-                else if (cur_char == '|') {
-                        /* Pipes are tokens that are delimiters and should be output
-                         * as a separate token
+                else if (next_char == '|') {
+                        /* Pipes are tokens that are delimiters and should be
+                         * output as a separate token
                          */
                         if (tok_pending_chars(tok)) {
@@ -128 +128 @@
                         tok_start_token(tok, TOKTYPE_PIPE);
                         
-                        rc = tok_push_char(tok, '|');
+                        rc = tok_push_char(tok, tok_get_char(tok));
                         if (rc != EOK) {
                                 return rc;
@@ -138 +138 @@
                         }
                 }
-                else if (cur_char == '\'') {
+                else if (next_char == '\'') {
                         /* A string starts with a quote (') and ends again with a quote.
                          * A literal quote is written as ''
                          */
                         tok_start_token(tok, TOKTYPE_TEXT);
+                        /* Eat the quote */
+                        tok_get_char(tok);
                         rc = tok_finish_string(tok);
                         if (rc != EOK) {
@@ -155 +157 @@
                          * the current token.
                          */
-                        rc = tok_push_char(tok, cur_char);
+                        rc = tok_push_char(tok, tok_get_char(tok));
                         if (rc != EOK) {
                                 return rc;
@@ -179 +181 @@
 {
         int rc;
-        wchar_t cur_char;
-        
-        while ((cur_char = tok_get_char(tok)) != 0) {
-                if (cur_char == '\'') {
+        wchar_t next_char;
+        
+        while ((next_char = tok_look_char(tok)) != 0) {
+                if (next_char == '\'') {
+                        /* Eat the quote */
+                        tok_get_char(tok);
                         if (tok_look_char(tok) == '\'') {
                                 /* Encode a single literal quote */
@@ -199 +203 @@
                 }
                 else {
-                        rc = tok_push_char(tok, cur_char);
+                        rc = tok_push_char(tok, tok_get_char(tok));
                         if (rc != EOK) {
                                 return rc;
@@ -255 +259 @@
         tokinfo->text = tok->outbuf + tok->outbuf_last_start;
         tokinfo->byte_start = tok->last_in_offset;
-        tokinfo->byte_length = tok->in_offset - tok->last_in_offset - 1;
+        tokinfo->byte_length = tok->in_offset - tok->last_in_offset;
         tokinfo->char_start = tok->last_in_char_offset;
-        tokinfo->char_length = tok->in_char_offset - tok->last_in_char_offset
-            - 1;
+        tokinfo->char_length = tok->in_char_offset - tok->last_in_char_offset;
         tok->outbuf_last_start = tok->outbuf_offset;
         
         /* We have consumed the first char of the next token already */
-        tok->last_in_offset = tok->in_offset-1;
-        tok->last_in_char_offset = tok->in_char_offset-1;
+        tok->last_in_offset = tok->in_offset;
+        tok->last_in_char_offset = tok->in_char_offset;
         
         return EOK;