Skip to content

Speed up creating Mark objects in the Cython version #848

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
bdraco wants to merge 6 commits into
base: main
Choose a base branch
from
Open

Speed up creating Mark objects in the Cython version #848

Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
64e40d8
Speed up creating Mark objects in the Cython version
bdraco Feb 21, 2025
964c641
Add `_create_mark` helper
bdraco May 16, 2025
688ebe7
avoid all the ref counting
bdraco May 16, 2025
0848153
Reduce diff size
bdraco May 16, 2025
4192cda
remove whitespace
bdraco May 16, 2025
aa051e8
document
bdraco May 16, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
46 changes: 30 additions & 16 deletions yaml/_yaml.pyx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -83,6 +83,20 @@ cdef class Mark:
% (self.name, self.line+1, self.column+1)
return where

# Helper function to create a Mark object with a single call
# This version does not have to convert size_t back to Python
# which makes it quite a bit faster when called from Cython
cdef Mark _create_mark(object name, size_t index, size_t line, size_t column,
object buffer, object pointer):
cdef Mark mark = Mark.__new__(Mark)
mark.name = name
mark.index = index
mark.line = line
mark.column = column
mark.buffer = buffer
mark.pointer = pointer
return mark

#class YAMLError(Exception):
# pass
#
Expand Down Expand Up @@ -310,12 +324,12 @@ cdef class CParser:
context_mark = None
problem_mark = None
if self.parser.context != NULL:
context_mark = Mark(self.stream_name,
context_mark = _create_mark(self.stream_name,
self.parser.context_mark.index,
self.parser.context_mark.line,
self.parser.context_mark.column, None, None)
if self.parser.problem != NULL:
problem_mark = Mark(self.stream_name,
problem_mark = _create_mark(self.stream_name,
self.parser.problem_mark.index,
self.parser.problem_mark.line,
self.parser.problem_mark.column, None, None)
Expand Down Expand Up @@ -356,12 +370,12 @@ cdef class CParser:
return token_object

cdef object _token_to_object(self, yaml_token_t *token):
start_mark = Mark(self.stream_name,
start_mark = _create_mark(self.stream_name,
token.start_mark.index,
token.start_mark.line,
token.start_mark.column,
None, None)
end_mark = Mark(self.stream_name,
end_mark = _create_mark(self.stream_name,
token.end_mark.index,
token.end_mark.line,
token.end_mark.column,
Expand Down Expand Up @@ -503,12 +517,12 @@ cdef class CParser:

cdef object _event_to_object(self, yaml_event_t *event):
cdef yaml_tag_directive_t *tag_directive
start_mark = Mark(self.stream_name,
start_mark = _create_mark(self.stream_name,
event.start_mark.index,
event.start_mark.line,
event.start_mark.column,
None, None)
end_mark = Mark(self.stream_name,
end_mark = _create_mark(self.stream_name,
event.end_mark.index,
event.end_mark.line,
event.end_mark.column,
Expand Down Expand Up @@ -673,7 +687,7 @@ cdef class CParser:
document = self._compose_document()
self._parse_next_event()
if self.parsed_event.type != YAML_STREAM_END_EVENT:
mark = Mark(self.stream_name,
mark = _create_mark(self.stream_name,
self.parsed_event.start_mark.index,
self.parsed_event.start_mark.line,
self.parsed_event.start_mark.column,
Expand All @@ -695,7 +709,7 @@ cdef class CParser:
if self.parsed_event.type == YAML_ALIAS_EVENT:
anchor = PyUnicode_FromYamlString(self.parsed_event.data.alias.anchor)
if anchor not in self.anchors:
mark = Mark(self.stream_name,
mark = _create_mark(self.stream_name,
self.parsed_event.start_mark.index,
self.parsed_event.start_mark.line,
self.parsed_event.start_mark.column,
Expand All @@ -715,7 +729,7 @@ cdef class CParser:
anchor = PyUnicode_FromYamlString(self.parsed_event.data.mapping_start.anchor)
if anchor is not None:
if anchor in self.anchors:
mark = Mark(self.stream_name,
mark = _create_mark(self.stream_name,
self.parsed_event.start_mark.index,
self.parsed_event.start_mark.line,
self.parsed_event.start_mark.column,
Expand All @@ -733,12 +747,12 @@ cdef class CParser:
return node

cdef _compose_scalar_node(self, object anchor):
start_mark = Mark(self.stream_name,
start_mark = _create_mark(self.stream_name,
self.parsed_event.start_mark.index,
self.parsed_event.start_mark.line,
self.parsed_event.start_mark.column,
None, None)
end_mark = Mark(self.stream_name,
end_mark = _create_mark(self.stream_name,
self.parsed_event.end_mark.index,
self.parsed_event.end_mark.line,
self.parsed_event.end_mark.column,
Expand Down Expand Up @@ -776,7 +790,7 @@ cdef class CParser:

cdef _compose_sequence_node(self, object anchor):
cdef int index
start_mark = Mark(self.stream_name,
start_mark = _create_mark(self.stream_name,
self.parsed_event.start_mark.index,
self.parsed_event.start_mark.line,
self.parsed_event.start_mark.column,
Expand Down Expand Up @@ -806,7 +820,7 @@ cdef class CParser:
value.append(self._compose_node(node, index))
index = index+1
self._parse_next_event()
node.end_mark = Mark(self.stream_name,
node.end_mark = _create_mark(self.stream_name,
self.parsed_event.end_mark.index,
self.parsed_event.end_mark.line,
self.parsed_event.end_mark.column,
Expand All @@ -815,7 +829,7 @@ cdef class CParser:
return node

cdef _compose_mapping_node(self, object anchor):
start_mark = Mark(self.stream_name,
start_mark = _create_mark(self.stream_name,
self.parsed_event.start_mark.index,
self.parsed_event.start_mark.line,
self.parsed_event.start_mark.column,
Expand Down Expand Up @@ -845,7 +859,7 @@ cdef class CParser:
item_value = self._compose_node(node, item_key)
value.append((item_key, item_value))
self._parse_next_event()
node.end_mark = Mark(self.stream_name,
node.end_mark = _create_mark(self.stream_name,
self.parsed_event.end_mark.index,
self.parsed_event.end_mark.line,
self.parsed_event.end_mark.column,
Expand Down Expand Up @@ -912,7 +926,7 @@ cdef class CEmitter:
if hasattr(stream, u'encoding'):
self.dump_unicode = 1
self.use_encoding = encoding
yaml_emitter_set_output(&self.emitter, output_handler, <void *>self)
yaml_emitter_set_output(&self.emitter, output_handler, <void *>self)
if canonical:
yaml_emitter_set_canonical(&self.emitter, 1)
if indent is not None:
Expand Down