In order to use filename_effective without spending bandwidth,
I have long performed a trial execution of curl directed
to a read=only directory, ignoring errors. This continues
to work, even for other resources on this problem site,
which was updated very recently. But today:
ifcurl 34+ chmod 555 /tmp/Darwin-... /Readonly
ifcurl 35+ : '========== Probe for FILE=%{filename_effective}'
iifcurl 42+ curl --verbose --location --header 'X-From: [email protected]'
--silent --show-error --output-dir /tmp/Darwin-... /Readonly --write-out
'%{filename_effective}' --remote-name --remote-header-name
'https://www.ibm.com/docs/en/zos/3.2.0?topic=SSLTBW_3.2.0/com.ibm.zos.v3r2.pdf/allpubs.htm'
* Host www.ibm.com:443 was resolved.
...
{ [281 bytes data]
< HTTP/2 200
...
{ [5 bytes data]
* client returned ERROR on write of 16375 bytes
{ [5 bytes data]
* Connection #0 to host www.ibm.com:443 left intact
curl: (23) client returned ERROR on write of 16375 bytes
ifcurl 42+ FILE=/tmp/Darwin-... /Readonly/3.2.0
ifcurl 44+ : 'Returned FILE='\''/tmp/Darwin-... /Readonly/3.2.0'\'''
(There's an implied wish here, that filename_effective
might return a useful result even with headers_only.
I want to save ETag with a related filename.)
--
Thanks,
gil
--
Unsubscribe: https://lists.haxx.se/mailman/listinfo/curl-users
Etiquette: https://curl.se/mail/etiquette.html
