Compare commits

..

2 Commits

Author SHA1 Message Date
msramalho
ef5b39c4f1 dind exception 2024-02-22 18:05:56 +00:00
msramalho
24ceafcb64 missing forward slash 2024-02-22 17:47:13 +00:00
2 changed files with 9 additions and 4 deletions

View File

@@ -40,7 +40,7 @@ class WaczArchiverEnricher(Enricher, Archiver):
self.use_docker = os.environ.get('WACZ_ENABLE_DOCKER') or not os.environ.get('RUNNING_IN_DOCKER')
self.docker_in_docker = os.environ.get('WACZ_ENABLE_DOCKER') and os.environ.get('RUNNING_IN_DOCKER')
self.cwd_dind = f"crawls/crawls{random_str(8)}"
self.cwd_dind = f"/crawls/crawls{random_str(8)}"
self.browsertrix_home_host = os.environ.get('BROWSERTRIX_HOME_HOST')
self.browsertrix_home_container = os.environ.get('BROWSERTRIX_HOME_CONTAINER') or self.browsertrix_home_host
# create crawls folder if not exists, so it can be safely removed in cleanup
@@ -120,7 +120,10 @@ class WaczArchiverEnricher(Enricher, Archiver):
logger.error(f"WACZ generation failed: {e}")
return False
if self.use_docker:
if self.docker_in_docker:
wacz_fn = os.path.join(self.cwd_dind, "collections", collection, f"{collection}.wacz")
elif self.use_docker:
wacz_fn = os.path.join(browsertrix_home_container, "collections", collection, f"{collection}.wacz")
else:
wacz_fn = os.path.join("collections", collection, f"{collection}.wacz")
@@ -133,7 +136,9 @@ class WaczArchiverEnricher(Enricher, Archiver):
if self.extract_media or self.extract_screenshot:
self.extract_media_from_wacz(to_enrich, wacz_fn)
if self.use_docker:
if self.docker_in_docker:
jsonl_fn = os.path.join(self.cwd_dind, "collections", collection, "pages", "pages.jsonl")
elif self.use_docker:
jsonl_fn = os.path.join(browsertrix_home_container, "collections", collection, "pages", "pages.jsonl")
else:
jsonl_fn = os.path.join("collections", collection, "pages", "pages.jsonl")

View File

@@ -3,7 +3,7 @@ _MAJOR = "0"
_MINOR = "9"
# On main and in a nightly release the patch should be one ahead of the last
# released build.
_PATCH = "5"
_PATCH = "7"
# This is mainly for nightly builds which have the suffix ".dev$DATE". See
# https://semver.org/#is-v123-a-semantic-version for the semantics.
_SUFFIX = ""