-
Notifications
You must be signed in to change notification settings - Fork 4k
ARROW-16818: [Doc][Python] Document GCS filesystem for PyArrow #13681
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.
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
Conversation
|
Also cc @wjones127 |
|
Based on my experience with the R docs, here a few specific things we should note:
|
|
Thanks for the suggestions @wjones127! I've added retry limit, anonymous connection and recursive parameters to the example, but didn't specifically comment on it. @emkornfield already wrote the docs, this mostly enables building them, with the exception of adding this example: This now renders: And: |
|
Please note - while I was able to build with This looks like another abseil issue but is probably MacOS/my laptop -specific. Has anyone else experienced this? I was able to locally build docs on a nightly build, so it's probably just me. |
|
I also had issues building with GCS on, is failing on the C++ side already (M1). Will keep trying. |
Co-authored-by: Ian Cook <ianmcook@gmail.com>
|
@coryan @emkornfield do you want to verify the abseil issue is limited to MacOS before release is cut (Monday)? |
|
Benchmark runs are scheduled for baseline = 38b956f and contender = 32016b1. 32016b1 is a master commit associated with this PR. Results will be available as each benchmark for each run completes. |
|
['Python', 'R'] benchmarks have high level of regressions. |
|
I think @lidavidm might have been done some upgrade on gRPC/absl after the GCS checkin. I was pretty sure we ran nightlies on mac, but maybe those weren't on M1. |
|
@rok what was the cmake command, and was abseil/gcs/etc. bundled, from conda, from homebrew…? |
|
I'm out at the moment, will check later today. Thanks for the input! |
|
I'm using the following cmake: All my dependencies (abseil/gcs/..) are installed via homebrew ( If I cmake with Otherwise compilation runs fine and things only fail when I import in Python or R. |
|
I suppose if we can figure out which abseil library is supposed to have that symbol, we can add it to the list of abseil libraries to link against in ThirdpartyToolchain.cmake. I don't have a Mac, and I don't exactly want to screw around with the build machines to debug this (since I usually SSH into those for reproducing Mac issues), so it'll be a little hard for me to help here 🙁 |
|
@lidavidm thanks for thinking about it though! 😄 |
|
I'm still not able to build GCS + tests, however GCS without tests builds fine and gives me a working build. This is the cmake (GCS + tests) that's erroring: Error from building GCS + tests: I've also noticed some duplicated symbols: These can be resolved with: diff --git a/cpp/cmake_modules/ThirdpartyToolchain.cmake b/cpp/cmake_modules/ThirdpartyToolchain.cmake
index 5d1da18b7..c9bbcf3db 100644
--- a/cpp/cmake_modules/ThirdpartyToolchain.cmake
+++ b/cpp/cmake_modules/ThirdpartyToolchain.cmake
@@ -4164,7 +4164,6 @@ macro(build_google_cloud_cpp_storage)
absl::variant
nlohmann_json::nlohmann_json
Crc32c::crc32c
- CURL::libcurl
Threads::Threads
OpenSSL::SSL
OpenSSL::Crypto
@@ -4193,9 +4192,7 @@ macro(build_google_cloud_cpp_storage)
absl::raw_logging_internal
absl::spinlock_wait
absl::strings
- absl::strings_internal
absl::str_format_internal
- absl::throw_delegate
absl::time
absl::time_zone
Crc32c::crc32c)I'm not sure how much of a problem this really is, but I'll open a Jira to make it more visible. |



This is to resolve ARROW-16818.