Repository navigation
Conversation
Add planarity.read(path), which returns a new PGraph populated from a graph file, and PGraph.read(path), which replaces the contents of an existing graph. Both read into a temporary C-layer graph and swap it in only after a successful read, so a failed read leaves the existing graph unchanged. Supports the adjacency list, adjacency matrix, graph6, and LEDA formats; GraphML remains write-only. Vertex labels are the C reader's vertex indexes, so one-based and zero-based inputs present the same labels, while a zero-based file still writes back zero-based. Also adds the cimport cython that Cython needs to recognize the @cython.embedsignature(False) decorator that keeps the non-public constructor out of the generated API docs.
_g6_ReadGraphFromStrOrFile() logged the failure from g6_ReadGraph() and then returned OK, so gp_Read() reported success for a malformed .g6 line and handed back an empty graph sized to the order the file declared. Return the reader's status instead, so callers can tell that the read failed. The Express read() API depends on this: its contract is that a failed read leaves the existing graph unchanged. Regression coverage for this lands with the Express read() changes in the following commit.
…ilure paths _from_path() allocated the C graph before encoding the path and before creating the Python object that owns it, so a path that raised on encoding left the graph allocated with nothing to free it. Encode the path first and create the owner before the allocation, so no fallible step sits between gp_New() and ownership. Also document that a path that cannot be encoded raises ValueError. Extend tests/test_express_api_read.py with invalid graph6 through both read entry points, which exercises the read-iterator fix in the previous commit, plus the exception paths that run before any allocation and a peak-RSS guard against the leak.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Resolves #79
Adds
planarity.read(path)to create a graph from a file andPGraph.read(path)to replace an existing graph.Failed reads leave the existing graph unchanged.
Type of change
Changes
This also fixes a graph6 reader bug: malformed input could report success and silently replace the graph.
The fix propagates the read failure, so it also affects
Graph.gp_Read()in the full API.I can split that fix into a separate PR if preferred.
Vertex labels
Read graphs use labels
1..N, including for zero-based input.The reader preserves the zero-based output flag, so writing the graph back can use labels starting at
0.Would you prefer the in-memory labels to follow the input's numbering instead?
Testing
pytest -q: 92 passed on CPython 3.14.7 and 3.11.8.These results cover the submitted branch at
c75cc5e.Integration with the latest
masterhas not yet been tested.