On 20/11/2015 19:23, Eric Blake wrote: > On 11/20/2015 02:04 AM, Paolo Bonzini wrote: >> JSONLexer only needs a simple resizable buffer. json-streamer.c >> can allocate memory for each token instead of relying on reference >> counting of QStrings. >> >> Signed-off-by: Paolo Bonzini <pbonz...@redhat.com> >> --- >> include/qapi/qmp/json-lexer.h | 7 +++---- >> include/qapi/qmp/json-streamer.h | 1 + >> qobject/json-lexer.c | 22 ++++++++-------------- >> qobject/json-streamer.c | 10 +++++----- >> 4 files changed, 17 insertions(+), 23 deletions(-) >> > >> --- a/include/qapi/qmp/json-streamer.h >> +++ b/include/qapi/qmp/json-streamer.h >> @@ -14,6 +14,7 @@ >> #ifndef QEMU_JSON_STREAMER_H >> #define QEMU_JSON_STREAMER_H >> >> +#include <stdint.h> >> #include "qapi/qmp/qlist.h" >> #include "qapi/qmp/json-lexer.h" >> > > Spurious hunk? If not, should mention in the commit message why this is > needed.
Does "because otherwise it doesn't compile" count? :) It was relying on json-lexer.h indirectly including stdint.h. I don't think it ought to be mentioned in the commit message. Paolo >> +++ b/qobject/json-streamer.c >> @@ -12,6 +12,7 @@ >> */ >> >> #include "qapi/qmp/qlist.h" >> +#include "qapi/qmp/qstring.h" >> #include "qapi/qmp/qint.h" >> #include "qapi/qmp/qdict.h" >> #include "qemu-common.h" >> @@ -21,13 +22,13 @@ >> #define MAX_TOKEN_SIZE (64ULL << 20) >> #define MAX_NESTING (1ULL << 10) >> >> -static void json_message_process_token(JSONLexer *lexer, QString *token, >> JSONTokenType type, int x, int y) >> +static void json_message_process_token(JSONLexer *lexer, GString *input, >> JSONTokenType type, int x, int y) > > Worth wrapping the long line while touching it? > > Otherwise looks like a straightforward conversion. If the commit > message is fixed, you can add > > Reviewed-by: Eric Blake <ebl...@redhat.com> >
