[Pkg-javascript-devel] Bug#1070940: node-follow-redirects: FTBFS in bullseye
Santiago Vila
sanvila at debian.org
Sat May 11 20:52:00 BST 2024
Package: src:node-follow-redirects
Version: 1.13.1-1+deb11u1
Severity: serious
Control: close -1 1.15.2+~1.14.1-1
Tags: ftbfs bullseye
Dear maintainer:
During a rebuild of all packages in bullseye, your package failed to build:
--------------------------------------------------------------------------------
[...]
debian/rules binary
dh binary
dh_update_autotools_config
dh_autoreconf
dh_auto_configure --buildsystem=nodejs
dh_auto_build --buildsystem=nodejs
No build command found, searching known files
dh_auto_test --buildsystem=nodejs
mkdir -p node_modules
ln -s ../. node_modules/follow-redirects
/bin/sh -ex debian/tests/pkg-js/test
+ mocha
follow-redirects
✓ http.get with string and callback - redirect (41ms)
✓ http.get with URL object and callback - redirect
✓ http.get with options object and callback - redirect
✓ http.get with string and callback - no redirect
✓ http.get with options object and callback - no redirect
✓ http.get with host option and callback - redirect
✓ http.get with response event
✓ should return with the original status code if the response does not contain a location header
✓ should emit connection errors on the returned stream
✓ should emit socket events on the returned stream
✓ should emit connect events on the returned stream
1) emits an error on redirects with an invalid location
✓ should destroy responses
✓ should honor query params in redirects
✓ should allow aborting
✓ should provide connection
✓ should provide flushHeaders
✓ should provide getHeader
✓ should provide removeHeader
✓ should provide setHeader
✓ should provide setNoDelay
✓ should provide setSocketKeepAlive
✓ should provide setTimeout
✓ should provide socket
✓ should wait for an explicit call to end
✓ errors on write after end
✓ should support writing into request stream without redirects
✓ should support writing into request stream with redirects
✓ should support piping into request stream without redirects
✓ should support piping into request stream with redirects
✓ should support piping into request stream with explicit Content-Length without redirects
✓ should support piping into request stream with explicit Content-Length with redirects
setTimeout
✓ clears timeouts after a successful response
✓ clears timeouts after an error response
✓ sets a timeout when the socket already exists
✓ should timeout on the final request
✓ should include redirect delays in the timeout
✓ overrides existing timeouts
should obey a `maxRedirects` property
✓ which defaults to 21 (52ms)
✓ which can be set globally
✓ set as an option on an individual request
the trackRedirects option
when not set
✓ should not track redirects
when set to true
✓ should track redirects
when the client passes an header named Authorization
✓ ignores it when null
when the client passes an header named Cookie
✓ ignores it when null
should switch to safe methods when appropriate
when redirecting with status code 300
✓ should reuse GET
✓ should reuse HEAD
✓ should reuse POST
✓ should reuse PUT
✓ should reuse DELETE
when redirecting with status code 301
✓ should reuse GET
✓ should reuse HEAD
✓ should switch from POST to GET
✓ should reuse PUT
✓ should reuse DELETE
when redirecting with status code 302
✓ should reuse GET
✓ should reuse HEAD
✓ should switch from POST to GET
✓ should reuse PUT
✓ should reuse DELETE
when redirecting with status code 303
✓ should reuse GET
✓ should reuse HEAD
✓ should switch from POST to GET
✓ should switch from PUT to GET
✓ should switch from DELETE to GET
when redirecting with status code 307
✓ should reuse GET
✓ should reuse HEAD
✓ should reuse POST
✓ should reuse PUT
✓ should reuse DELETE
should error on an unsupported protocol redirect
✓ (http -> about)
should obey a `maxBodyLength` property
✓ which defaults to 10MB
✓ set globally, on write
✓ set per request, on write
✓ set globally, on end
✓ set per request, on end
writing invalid data
✓ throws an error
when switching from POST to GET
✓ should drop the entity and associated headers
when redirecting to a different host while the host header is set
✓ uses the new host header
when the client passes an Authorization header
✓ keeps the header when redirected to the same host
✓ keeps the header when redirected to the same host via header
✓ drops the header when redirected to a different host
✓ drops the header when redirected from a different host via header
when the followRedirects option is set to false
✓ does not redirect
should not hang on empty writes
✓ when data is the empty string without encoding
✓ when data is the empty string with encoding
✓ when data is Buffer.from('')
end accepts as arguments
✓ (none)
✓ the empty string
✓ a non-empty string
✓ a non-empty string and an encoding
✓ a non-empty string, an encoding, and a callback
✓ a non-empty string and a callback
✓ a callback
change request options before redirects
✓ only call beforeRedirect on redirects, not the inital http call
✓ ignore beforeRedirect if not a function
✓ append new header with every redirect
✓ abort request chain after throwing an error
✓ access response header in beforeRedirect
98 passing (416ms)
1 failing
1) follow-redirects
emits an error on redirects with an invalid location:
AssertionError [ERR_ASSERTION]: 'HPE_INVALID_HEADER_TOKEN' == 'ERR_FR_REDIRECTION_FAILURE'
+ expected - actual
-HPE_INVALID_HEADER_TOKEN
+ERR_FR_REDIRECTION_FAILURE
at /<<PKGBUILDDIR>>/test/test.js:283:16
at processTicksAndRejections (internal/process/task_queues.js:97:5)
dh_auto_test: error: /bin/sh -ex debian/tests/pkg-js/test returned exit code 1
make: *** [debian/rules:8: binary] Error 25
dpkg-buildpackage: error: debian/rules binary subprocess returned exit status 2
--------------------------------------------------------------------------------
The above is just how the build ends and not necessarily the most relevant part.
If required, the full build log is available here:
https://people.debian.org/~sanvila/build-logs/bullseye/
About the archive rebuild: The build was made on virtual machines
of type m6a.large and r6a.large from AWS, using sbuild and a
reduced chroot with only build-essential packages.
If you could not reproduce the bug please contact me privately, as I
am willing to provide ssh access to a virtual machine where the bug is
fully reproducible.
If this is really a bug in one of the build-depends, please use
reassign and affects, so that this is still visible in the BTS web
page for this package.
Thanks.
More information about the Pkg-javascript-devel
mailing list