Add HTMLRewriter class in Python SDK #5945
Conversation
| wrapped_body = js.ReadableStream.new( | ||
| to_js( | ||
| {"start": start_proxy}, | ||
| dict_converter=Object.fromEntries, | ||
| ) | ||
| ) |
There was a problem hiding this comment.
This is basically the main job for this wrapper class, but it only provides start for now.
I considered adding cancel as well, but using cancel actually makes the proxy lifetime handling a lot much complicated as I found that even after calling cancel, there might be remaining task that tries to read the stream, and destroying the proxy too early would crash the code...
| # FIXME: The exception is caught correctly in Python-level, | ||
| # but workerd shows "uncaught exception" error which makes the test fail. | ||
| # await test_handler_error() |
There was a problem hiding this comment.
Passing a Python class that raise exception to the JS HTMLRewriter always makes the test fail even if I catch the error in Python (it is catched in the Python block but the test still fails). I am not sure if it is expected behavior or not.
workerd/server/server.c++:5882: debug: [ TEST ] python-htmlrewriter
workerd/io/worker.c++:2323: info: uncaught exception; source = Uncaught; stack = PythonError: Traceback (most recent call last):
File "/session/metadata/worker.py", line 202, in text
raise ValueError("Test error")
ValueError: Test error
at new_error (pyodide-internal:generated/emscriptenSetup:19907:14)
at wasm://wasm/020fb45e:wasm-function[261]:0x10f02c
at wasm://wasm/020fb45e:wasm-function[262]:0x10f0f4
at callPyObjectKwargs (pyodide-internal:generated/emscriptenSetup:22399:37)
at Module.callPyObject (pyodide-internal:generated/emscriptenSetup:22444:18)
at Function.apply (pyodide-internal:generated/emscriptenSetup:23275:60)
at Object.apply (pyodide-internal:generated/emscriptenSetup:23118:20)
workerd/io/io-context.c++:430: info: uncaught exception; exception = workerd/jsg/_virtual_includes/iterator/workerd/jsg/value.h:1480: failed: jsg.Error: PythonError: Traceback (most recent call last):
File "/session/metadata/worker.py", line 202, in text
raise ValueError("Test error")
ValueError: Test error
| to_js( | ||
| {"start": start_proxy}, | ||
| dict_converter=Object.fromEntries, | ||
| ) |
There was a problem hiding this comment.
| to_js( | |
| {"start": start_proxy}, | |
| dict_converter=Object.fromEntries, | |
| ) | |
| start=start_proxy |
There was a problem hiding this comment.
The constructor takes an object (https://developer.mozilla.org/en-US/docs/Web/API/ReadableStream/ReadableStream#underlyingsource)
ryanking13
force-pushed
the
gyeongjae/htmlrewriter-2
branch
from
d71bcaa to
41c35e0
Compare
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## main #5945 +/- ##
===========================================
+ Coverage 68.10% 69.67% +1.56%
===========================================
Files 29 397 +368
Lines 2593 105999 +103406
Branches 15 17965 +17950
===========================================
+ Hits 1766 73853 +72087
- Misses 827 21351 +20524
- Partials 0 10795 +10795 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
3 participants
This adds
HTMLRewriterclass in Python SDK, which does proxy lifetime handling internally.We have an example that uses HTMLRewriter, but it uses JS API directly which is slightly annoying to use in Python because of the proxy handling.