I think it would be best if LUCI could offer a link directly to download the text only log. Without being able to grep through a text file or use it in vim it is easy to miss problems.

RE ihf@, it's not as nice as a straight text-only download service, but you could always get the logs in text form using console:

Prerequisite: "depot_tools" in PATH.

$ cit logdog cat chromeos/bb/.../path/+/to/log/stdout > log.txt

Doesn't work for me.

Example log
https://logs.chromium.org/v/?s=chromeos%2Fbb%2Fchromeos%2Fcaroline-chrome-pfq%2F1889%2F%2B%2Frecipes%2Fsteps%2FSyncChrome%2F0%2Fstdout

Failure
ihf@ql:~$ cit logdog cat chromeos%2Fbb%2Fchromeos%2Fcaroline-chrome-pfq%2F1889%2F%2B%2Frecipes%2Fsteps%2FSyncChrome%2F0%2Fstdout > log.txt
[E2018-04-27T10:25:51.947311-07:00 236321 0 subcommandCat.go:108] Invalid path specifier.                     {"error":"invalid project name \"chromeos%2Fbb%2Fchromeos%2Fcaroline-chrome-pfq%2F1889%2F%2B%2Frecipes%2Fsteps%2FSyncChrome%2F0%2Fstdout\": invalid character at 8 (%)"}

try:

cit logdog cat chromeos/bb/chromeos/caroline-chrome-pfq/1889/+/recipes/steps/SyncChrome/0/stdout

So, my point is I need something that is
a) simple to use
b) efficient/quick

I can't manually replace all the special characters. I have to use this dozens of times a day.

I just copied this string here: https://screenshot.googleplex.com/w7Zi45RbfFN

ihf@ql:~$ cit logdog cat chromeos/bb/chromeos/caroline-chrome-pfq/1889/+/recipes/steps/SyncChrome/0/stdout
[W2018-04-27T13:21:34.657556-07:00 25249 0 client.go:304] RPC failed permanently: rpc error: code = Unauthenticated desc =  {"error":"rpc error: code = Unauthenticated desc = ", "host":"luci-logdog.appspot.com", "method":"Get", "service":"logdog.Logs"}
[E2018-04-27T13:21:34.657669-07:00 25249 0 fetcher.go:343] Fetch returned error.                       {"error":"coordinator: no access", "index":0}
[E2018-04-27T13:21:34.657734-07:00 25249 0 fetcher.go:281] Error fetching logs.                        {"error":"coordinator: no access"}
[E2018-04-27T13:21:34.657819-07:00 25249 0 subcommandCat.go:160] Failed to fetch log stream.                 {"error":"coordinator: no access", "index":0, "path":"bb/chromeos/caroline-chrome-pfq/1889/+/recipes/steps/SyncChrome/0/stdout", "project":"chromeos"}

I have prodaccess at the moment, if that matters.

The log is non-public, so you need to run:

"cit logdog auth-login"

That works. Now the downloads happen at about 1MB/s. Why so slow? I can reliably download a longer file from pantheon in *less* than 3 seconds, e.g. close to pipe speed of 1Gb/s:

LogDog:

ihf@ql:~$ time cit logdog cat chromeos/bb/chromeos/caroline-chrome-pfq/1889/+/recipes/steps/SyncChrome/0/stdout > log
real	0m5.435s
-rw-r--r-- 1 ihf primarygroup 3.8M Apr 27 13:27 log

ihf@ql:~$ time cit logdog cat chromeos/bb/chromeos/betty-incremental/562/+/recipes/steps/VMTest__attempt_1_/0/stdout > log
real	1m38.837s
-rw-r--r-- 1 ihf primarygroup 144M Apr 27 13:30 log


Pantheon/browser download:
https://pantheon.corp.google.com/storage/browser/chromeos-image-archive/betty-incremental/R68-10622.0.0-b2518654/vm_test_results_1/vmtest-informational4/test_harness/all/SimpleTestVerify/1_autotest_tests/results-1-cheets_CTS_N.vm.x86.CtsDeqpTestCases/debug/
-rw-r--r-- 1 ihf primarygroup 161M Apr 27 13:37 betty-incremental

Three ways to download logs now:

1. cit logdog cat ...
2. curl the URL
3. add ?format=raw to the URL

While we don't actually have a download link, ctrl + s will technically work too (but it won't be plaintext).

Thanks a lot!