git/list[1] front-page[2] threads[3] people[4] search[5] about
 

[PATCH 08/26] update-ref --stdin: Read the whole input at once

From
Michael Haggerty <mhagger@alum.mit.edu>
Date
Mar 10, 2014, 12:46 UTC
Message-ID
<1394455603-2968-9-git-send-email-mhagger@alum.mit.edu>
In-Reply-To
<1394455603-2968-1-git-send-email-mhagger@alum.mit.edu>

Decouple the parsing code from the input source (the old parsing code had to read new data even in the middle of commands). This might also be a tad faster, but that is inconsequential. Add docstrings for the parsing functions.

Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
---
 builtin/update-ref.c | 170 ++++++++++++++++++++++++++++++++-------------------
 1 file changed, 108 insertions(+), 62 deletions(-)
diff --git a/builtin/update-ref.c b/builtin/update-ref.c
index a8a68e8..5f197fe 100644
--- a/builtin/update-ref.c
+++ b/builtin/update-ref.c
@@ -85,44 +85,70 @@ static const char *parse_arg(const char *next, struct strbuf *arg)
 	return next;
 }
 
-static const char *parse_first_arg(const char *next, struct strbuf *arg)
+/*
+ * Parse the argument immediately after "command SP".  If not -z, then
+ * handle C-quoting.  Write the argument to arg.  Set *next to point
+ * at the character that terminates the argument.  Die if C-quoting is
+ * malformed.
+ */
+static void parse_first_arg(struct strbuf *input, const char **next,
+			    struct strbuf *arg)
 {
-	/* Parse argument immediately after "command SP" */
 	strbuf_reset(arg);
 	if (line_termination) {
 		/* Without -z, use the next argument */
-		next = parse_arg(next, arg);
+		*next = parse_arg(*next, arg);
 	} else {
-		/* With -z, use rest of first NUL-terminated line */
-		strbuf_addstr(arg, next);
-		next = next + arg->len;
+		/* With -z, use everything up to the next NUL */
+		strbuf_addstr(arg, *next);
+		*next += arg->len;
 	}
-	return next;
 }
 
-static const char *parse_next_arg(const char *next, struct strbuf *arg)
+/*
+ * Parse a SP/NUL separator followed by the next SP- or NUL-terminated
+ * argument, if any.  If there is an argument, write it to arg, set
+ * *next to point at the character terminating the argument, and
+ * return 0.  If there is no argument at all (not even the empty
+ * string), return a non-zero result and leave *next unchanged.
+ */
+static int parse_next_arg(struct strbuf *input, const char **next,
+			  struct strbuf *arg)
 {
-	/* Parse next SP-terminated or NUL-terminated argument, if any */
 	strbuf_reset(arg);
 	if (line_termination) {
 		/* Without -z, consume SP and use next argument */
-		if (!*next)
-			return NULL;
-		if (*next != ' ')
-			die("expected SP but got: %s", next);
-		next = parse_arg(next + 1, arg);
+		if (!**next || **next == line_termination)
+			return -1;
+		if (**next != ' ')
+			die("expected SP but got: %s", *next);
+		(*next)++;
+		*next = parse_arg(*next, arg);
 	} else {
 		/* With -z, read the next NUL-terminated line */
-		if (*next)
-			die("expected NUL but got: %s", next);
-		if (strbuf_getline(arg, stdin, '\0') == EOF)
-			return NULL;
-		next = arg->buf + arg->len;
+		if (**next)
+			die("expected NUL but got: %s", *next);
+		(*next)++;
+		if (*next == input->buf + input->len)
+			return -1;
+		strbuf_addstr(arg, *next);
+		*next += arg->len;
 	}
-	return next;
+	return 0;
 }
 
-static void parse_cmd_update(const char *next)
+
+/*
+ * The following five parse_cmd_*() functions parse the corresponding
+ * command.  In each case, next points at the character following the
+ * command name and the following space.  They each return a pointer
+ * to the character terminating the command, and die with an
+ * explanatory message if there are any parsing problems.  All of
+ * these functions handle either text or binary format input,
+ * depending on how line_termination is set.
+ */
+
+static const char *parse_cmd_update(struct strbuf *input, const char *next)
 {
 	struct strbuf ref = STRBUF_INIT;
 	struct strbuf newvalue = STRBUF_INIT;
@@ -131,26 +157,28 @@ static void parse_cmd_update(const char *next)
 
 	update = update_alloc();
 
-	if ((next = parse_first_arg(next, &ref)) != NULL && ref.buf[0])
+	parse_first_arg(input, &next, &ref);
+	if (ref.buf[0])
 		update_store_ref_name(update, ref.buf);
 	else
 		die("update line missing <ref>");
 
-	if ((next = parse_next_arg(next, &newvalue)) != NULL)
+	if (!parse_next_arg(input, &next, &newvalue))
 		update_store_new_sha1(update, newvalue.buf);
 	else
 		die("update %s missing <newvalue>", ref.buf);
 
-	if ((next = parse_next_arg(next, &oldvalue)) != NULL)
+	if (!parse_next_arg(input, &next, &oldvalue)) {
 		update_store_old_sha1(update, oldvalue.buf);
-	else if(!line_termination)
+		if (*next != line_termination)
+			die("update %s has extra input: %s", ref.buf, next);
+	} else if (!line_termination)
 		die("update %s missing [<oldvalue>] NUL", ref.buf);
 
-	if (next && *next)
-		die("update %s has extra input: %s", ref.buf, next);
+	return next;
 }
 
-static void parse_cmd_create(const char *next)
+static const char *parse_cmd_create(struct strbuf *input, const char *next)
 {
 	struct strbuf ref = STRBUF_INIT;
 	struct strbuf newvalue = STRBUF_INIT;
@@ -158,23 +186,27 @@ static void parse_cmd_create(const char *next)
 
 	update = update_alloc();
 
-	if ((next = parse_first_arg(next, &ref)) != NULL && ref.buf[0])
+	parse_first_arg(input, &next, &ref);
+	if (ref.buf[0])
 		update_store_ref_name(update, ref.buf);
 	else
 		die("create line missing <ref>");
 
-	if ((next = parse_next_arg(next, &newvalue)) != NULL)
+	if (!parse_next_arg(input, &next, &newvalue))
 		update_store_new_sha1(update, newvalue.buf);
 	else
 		die("create %s missing <newvalue>", ref.buf);
+
 	if (is_null_sha1(update->new_sha1))
 		die("create %s given zero new value", ref.buf);
 
-	if (next && *next)
+	if (*next != line_termination)
 		die("create %s has extra input: %s", ref.buf, next);
+
+	return next;
 }
 
-static void parse_cmd_delete(const char *next)
+static const char *parse_cmd_delete(struct strbuf *input, const char *next)
 {
 	struct strbuf ref = STRBUF_INIT;
 	struct strbuf oldvalue = STRBUF_INIT;
@@ -182,23 +214,26 @@ static void parse_cmd_delete(const char *next)
 
 	update = update_alloc();
 
-	if ((next = parse_first_arg(next, &ref)) != NULL && ref.buf[0])
+	parse_first_arg(input, &next, &ref);
+	if (ref.buf[0])
 		update_store_ref_name(update, ref.buf);
 	else
 		die("delete line missing <ref>");
 
-	if ((next = parse_next_arg(next, &oldvalue)) != NULL)
+	if (!parse_next_arg(input, &next, &oldvalue)) {
 		update_store_old_sha1(update, oldvalue.buf);
-	else if(!line_termination)
+		if (update->have_old && is_null_sha1(update->old_sha1))
+			die("delete %s given zero old value", ref.buf);
+	} else if (!line_termination)
 		die("delete %s missing [<oldvalue>] NUL", ref.buf);
-	if (update->have_old && is_null_sha1(update->old_sha1))
-		die("delete %s given zero old value", ref.buf);
 
-	if (next && *next)
+	if (*next != line_termination)
 		die("delete %s has extra input: %s", ref.buf, next);
+
+	return next;
 }
 
-static void parse_cmd_verify(const char *next)
+static const char *parse_cmd_verify(struct strbuf *input, const char *next)
 {
 	struct strbuf ref = STRBUF_INIT;
 	struct strbuf value = STRBUF_INIT;
@@ -206,53 +241,64 @@ static void parse_cmd_verify(const char *next)
 
 	update = update_alloc();
 
-	if ((next = parse_first_arg(next, &ref)) != NULL && ref.buf[0])
+	parse_first_arg(input, &next, &ref);
+	if (ref.buf[0])
 		update_store_ref_name(update, ref.buf);
 	else
 		die("verify line missing <ref>");
 
-	if ((next = parse_next_arg(next, &value)) != NULL) {
+	if (!parse_next_arg(input, &next, &value)) {
 		update_store_old_sha1(update, value.buf);
 		update_store_new_sha1(update, value.buf);
-	} else if(!line_termination)
+	} else if (!line_termination)
 		die("verify %s missing [<oldvalue>] NUL", ref.buf);
 
-	if (next && *next)
+	if (*next != line_termination)
 		die("verify %s has extra input: %s", ref.buf, next);
+
+	return next;
 }
 
-static void parse_cmd_option(const char *next)
+static const char *parse_cmd_option(struct strbuf *input, const char *next)
 {
-	if (!strcmp(next, "no-deref"))
+	if (!strncmp(next, "no-deref", 8) && next[8] == line_termination)
 		update_flags |= REF_NODEREF;
 	else
 		die("option unknown: %s", next);
+	return next + 8;
 }
 
 static void update_refs_stdin(void)
 {
-	struct strbuf cmd = STRBUF_INIT;
+	struct strbuf input = STRBUF_INIT;
+	const char *next;
 
+	if (strbuf_read(&input, 0, 1000) < 0)
+		die_errno("could not read from stdin");
+	next = input.buf;
 	/* Read each line dispatch its command */
-	while (strbuf_getline(&cmd, stdin, line_termination) != EOF)
-		if (!cmd.buf[0])
+	while (next < input.buf + input.len) {
+		if (*next == line_termination)
 			die("empty command in input");
-		else if (isspace(*cmd.buf))
-			die("whitespace before command: %s", cmd.buf);
-		else if (starts_with(cmd.buf, "update "))
-			parse_cmd_update(cmd.buf + 7);
-		else if (starts_with(cmd.buf, "create "))
-			parse_cmd_create(cmd.buf + 7);
-		else if (starts_with(cmd.buf, "delete "))
-			parse_cmd_delete(cmd.buf + 7);
-		else if (starts_with(cmd.buf, "verify "))
-			parse_cmd_verify(cmd.buf + 7);
-		else if (starts_with(cmd.buf, "option "))
-			parse_cmd_option(cmd.buf + 7);
+		else if (isspace(*next))
+			die("whitespace before command: %s", next);
+		else if (starts_with(next, "update "))
+			next = parse_cmd_update(&input, next + 7);
+		else if (starts_with(next, "create "))
+			next = parse_cmd_create(&input, next + 7);
+		else if (starts_with(next, "delete "))
+			next = parse_cmd_delete(&input, next + 7);
+		else if (starts_with(next, "verify "))
+			next = parse_cmd_verify(&input, next + 7);
+		else if (starts_with(next, "option "))
+			next = parse_cmd_option(&input, next + 7);
 		else
-			die("unknown command: %s", cmd.buf);
+			die("unknown command: %s", next);
+
+		next++;
+	}
 
-	strbuf_release(&cmd);
+	strbuf_release(&input);
 }
 
 int cmd_update_ref(int argc, const char **argv, const char *prefix)

-- 
1.9.0
Previous: Michael HaggertyNext: Michael Haggerty
Message 16 of 38 in “Clean up update-refs --stdin and implement ref_transaction”
  1. 00/26 Clean up update-refs --stdin and implement ref_transactionMichael Haggerty, Mar 10, 2014
  2. 01/26 t1400: Fix name and expected result of one testMichael Haggerty, Mar 10, 2014
  3. 02/26 t1400: Provide sensible input to the commandMichael Haggerty, Mar 10, 2014
  4. 03/26 t1400: Pass a legitimate <newvalue> to update commandMichael Haggerty, Mar 10, 2014
  5. Brad KingMar 10, 2014
  6. Michael HaggertyMar 10, 2014
  7. Brad KingMar 11, 2014
  8. Junio C HamanoMar 11, 2014
  9. Brad KingMar 11, 2014
  10. Michael HaggertyMar 20, 2014
  11. 04/26 parse_arg(): Really test that argument is properly terminatedMichael Haggerty, Mar 10, 2014
  12. 05/26 t1400: Add some more tests involving quoted argumentsMichael Haggerty, Mar 10, 2014
  13. Johan HerlandMar 10, 2014
  14. 06/26 refs.h: Rename the action_on_err constantsMichael Haggerty, Mar 10, 2014
  15. 07/26 update_refs(): Fix constnessMichael Haggerty, Mar 10, 2014
  16. 08/26 update-ref --stdin: Read the whole input at onceMichael Haggerty, Mar 10, 2014
  17. 09/26 parse_cmd_verify(): Copy old_sha1 instead of evaluating <oldvalue> twiceMichael Haggerty, Mar 10, 2014
  18. 10/26 update-ref.c: Extract a new function, parse_refname()Michael Haggerty, Mar 10, 2014
  19. 11/26 update-ref --stdin: Improve error messages for invalid valuesMichael Haggerty, Mar 10, 2014
  20. 12/26 update-ref --stdin: Make error messages more consistentMichael Haggerty, Mar 10, 2014
  21. 13/26 update-ref --stdin: Simplify error messages for missing oldvaluesMichael Haggerty, Mar 10, 2014
  22. Brad KingMar 10, 2014
  23. Brad KingMar 10, 2014
  24. 14/26 update-ref.c: Extract a new function, parse_next_sha1()Michael Haggerty, Mar 10, 2014
  25. 15/26 update-ref --stdin: Improve the error message for unexpected EOFMichael Haggerty, Mar 10, 2014
  26. 16/26 update-ref --stdin: Harmonize error messagesMichael Haggerty, Mar 10, 2014
  27. 17/26 refs: Add a concept of a reference transactionMichael Haggerty, Mar 10, 2014
  28. 18/26 update-ref --stdin: Reimplement using reference transactionsMichael Haggerty, Mar 10, 2014
  29. 19/26 refs: Remove API function update_refs()Michael Haggerty, Mar 10, 2014
  30. 20/26 struct ref_update: Rename field "ref_name" to "refname"Michael Haggerty, Mar 10, 2014
  31. 21/26 struct ref_update: Store refname as a FLEX_ARRAY.Michael Haggerty, Mar 10, 2014
  32. 22/26 commit_ref_transaction(): Introduce temporary variablesMichael Haggerty, Mar 10, 2014
  33. 23/26 struct ref_update: Add a lock memberMichael Haggerty, Mar 10, 2014
  34. 24/26 struct ref_update: Add type fieldMichael Haggerty, Mar 10, 2014
  35. 25/26 commit_ref_transaction(): Also free the ref_transactionMichael Haggerty, Mar 10, 2014
  36. 26/26 commit_ref_transaction(): Work with transaction->updates in placeMichael Haggerty, Mar 10, 2014
  37. Brad KingMar 10, 2014
  38. Michael HaggertyMar 10, 2014

Read the whole thread, see it on lore, or plain text.

$ cat FOOTERMessages come from the public archive at lore.kernel.org/git, fetched every hour. The front page is chosen and written each morning by an AI editor and can be wrong; the threads themselves are the record. About and API. For agents: an MCP server at https://gitlist.dev/mcp, and any thread, story or person page as Markdown by adding .md to its URL (or sending Accept: text/markdown). Details in /llms.txt.