10c332d5d2
libsysprof-analyze: index traceables for faster access
...
This allows us to skip past all the nodes we don't care about with
relatively low overhead once the document has been generated.
2023-05-03 17:44:15 -07:00
b72eda6be2
libsysprof-analyze: prepare symbolizers before decoding
2023-05-03 17:22:37 -07:00
95325cc26f
libsysprof-analyze: add API to load embedded file as gbytes
2023-05-03 17:07:47 -07:00
c0ade7da01
libsysprof-analyze: add helper for native byte order
2023-05-03 17:07:19 -07:00
426aaad781
libsysprof-analyze: add plumbing for symbolization
...
This gets the worker thread going and the state passed to eventually do
symbolization through the new symbolizer API.
2023-05-03 14:59:46 -07:00
4505e8bb85
libsysprof-analyze: use ref strings internally
...
This gives us a bit more control than the string chunk, so long as we can
rely on the allocator to do something nice.
2023-05-02 13:30:09 -07:00
ffb6533e02
libsysprof-analyze: add API for intern'ing strings in document
...
We will want to re-use strings as much as we can for resolving symbols,
tags, etc.
2023-05-01 11:39:55 -07:00
97c93ea965
libsysprof-analyze: track frame length
...
That way we don't have to decode this too in the subclasses.
2023-04-25 17:09:04 -07:00
a5dafa5409
libsysprof-analyze: rename some private API
...
prefix with _, use frame type directly, use needs_swap rather than
is_native so it's more clear to readers.
2023-04-25 16:11:59 -07:00
31003c519c
libsysprof-analyze: rename to SysprofDocumentFrame
...
We will eventually be adding sub-types for the various frame types, and
use this as a common ancestor for item inflation.
2023-04-25 15:32:55 -07:00
ed01673a5e
libsysprof-analyzer: rename model to SysprofDocument
...
This will provide better namespacing for the objects inflated from the
document for various frame types. By creating real objects with real
properties we give ourselves quite a bit of flexibility in the data
filtering language coming forth.
2023-04-25 15:26:06 -07:00
