Handle long messages in event parser

[aldld]

This PR fixes a bug that causes the go-sse client to drop messages that are longer than 4096 bytes with an Unexpected EOF error.

The issue here is that the Scanner that gets used to fill the input buffer for the byte parser (which gets used to extract fields) may not call the SplitFunc with a buffer containing a full token. In that case, the split function needs to request more data, signalling to the Scanner to call the SplitFunc again with a larger buffer.

This PR includes a new test case in TestReaderParser that fails if this change is not present, along with a new test in internal/parser/split_func_test.go.

[tmaxmax]

Thank you for raising the issue and the PR! The requested changes address only some style preferences. It's also great you've added tests for this edge case – seems like I missed that part of bufio.Scanner's behavior. I appreciate you taking time to find and fix this issue. Was it easy navigating the codebase? If you have any naming/layout/other suggestions, please let me know.

[tmaxmax]

Also make sure to rebase on master – I've updated and fixed the actions.

[aldld]

Thank you for raising the issue and the PR! The requested changes address only some style preferences. It's also great you've added tests for this edge case – seems like I missed that part of bufio.Scanner's behavior. I appreciate you taking time to find and fix this issue. Was it easy navigating the codebase? If you have any naming/layout/other suggestions, please let me know.

Thanks for the feedback! I've rebased and updated the PR based on your comments. I found most of the codebase quite easy to follow, particularly the top-level package and the readme documentation.

It did take me some time to wrap my head around the internal/parser package, mainly in terms of understanding the relationship between the different functions named Scan that get called by ReaderParser.Scan(). It was a bit difficult to keep in my head the difference between e.g. r.p.Scan() and r.sc.Scan(), so perhaps naming those fields a bit differently (maybe eventScanner and fieldScanner?) would help clarify what's happening at each step.

Other than that, though, I had no trouble understanding and navigating the codebase.

[tmaxmax]

Thank you for raising the issue and the PR! The requested changes address only some style preferences. It's also great you've added tests for this edge case – seems like I missed that part of bufio.Scanner's behavior. I appreciate you taking time to find and fix this issue. Was it easy navigating the codebase? If you have any naming/layout/other suggestions, please let me know.

Thanks for the feedback! I've rebased and updated the PR based on your comments. I found most of the codebase quite easy to follow, particularly the top-level package and the readme documentation.

It did take me some time to wrap my head around the internal/parser package, mainly in terms of understanding the relationship between the different functions named Scan that get called by ReaderParser.Scan(). It was a bit difficult to keep in my head the difference between e.g. r.p.Scan() and r.sc.Scan(), so perhaps naming those fields a bit differently (maybe eventScanner and fieldScanner?) would help clarify what's happening at each step.

Other than that, though, I had no trouble understanding and navigating the codebase.

That's a very useful suggestion, thank you! I'll implement it in another PR. Will merge your PR and release a new version soon, as I've found another bug and an opportunity to remove some code from the codebase.

[codecov-commenter]

Codecov Report

Patch coverage: 100.00% and project coverage change: +0.02 🎉

Comparison is base (3baa34d) 96.39% compared to head (c235a39) 96.41%.

❗ Your organization is not using the GitHub App Integration. As a result you may experience degraded service beginning May 15th. Please install the Github App Integration for your organization. Read more.

Additional details and impacted files

@@            Coverage Diff             @@
##           master       #2      +/-   ##
==========================================
+ Coverage   96.39%   96.41%   +0.02%     
==========================================
  Files          14       14              
  Lines        1054     1060       +6     
==========================================
+ Hits         1016     1022       +6     
  Misses         30       30              
  Partials        8        8              

Impacted Files	Coverage Δ
internal/parser/reader_parser.go	100.00% <100.00%> (ø)

☔ View full report in Codecov by Sentry.
📢 Do you have feedback about the report comment? Let us know in this issue.

[tmaxmax]

See latest release. I've also made the suggested refactor.