src/usr.bin/fetch/fetch.c, branch stable/14

fetch: correct env var name for --crl option to work

2024-10-09T11:49:32+00:00

Reviewed by:	grembo
MFC after:	3 days
Differential Revision: https://reviews.freebsd.org/D46973

(cherry picked from commit 1d7a33888e09f86feb4c2c9f9640d936be07ce9d)

MFC: fetch(1): suppress "Not Modified" in quiet mode

2024-09-14T07:21:07+00:00

"fetch -qi" should skip printing "Not Modified" for successful
http(s) request. Still print it by default (v_level == 1).

(cherry picked from commit 7a6309fdc79427b0a7a5c3876daba150d946ae22)

Remove $FreeBSD$: one-line .c pattern

2023-08-16T17:54:42+00:00

Remove /^[\s*]*__FBSDID\("\$FreeBSD\$"\);?\s*\n/

fetch: make -S argument accept values > 2GB

2022-02-20T14:24:43+00:00

Use strtoll() to parse the argument of the -S option.

FreeBSD has supported 64 bit file offsets for more than 25 years on
all architectures and off_t is a 64 bit integer type for that reason.

While strtol() returns a 64 bit value on 64 LP64 architectures, it
is limit to 32 bit on e.g. i386. The strtoll() function returns a 64
but result on all supported architectures and therefore supports the
possible file lengths and file offsets on 32 bit archtectures.

Reported by:	antoine
MFC after:	3 days

fetch(1): more fixes for soft failure handling

2022-01-25T05:48:28+00:00

Fix logic error introduced in my commit
bf599c03f09dea0f7e188e002b42d782af6841c3

Also, authorization errors should not be considered as soft failures.

fetch(1): fix error in previous commit

2022-01-24T08:07:18+00:00

strncmp() compares terminating zero and sizeof() includes it.
Un-obsfuscate the code and show what it is indended to do.

fetch(1): do not consider HTTP 5XX errors as soft failures

2022-01-24T07:38:26+00:00

This change fixes "fetch -a" looping forever on "502 Bad gateway"
error and similar.

MFC after:	1 month

fetch(1): correct progress accounting after previous commit

2022-01-24T04:17:24+00:00

MFC after:	1 month

fetch(1): process truncated transfer as soft failure

2022-01-24T04:09:37+00:00

Let "fetch -a" resume truncated transfer automatically
perform another attempt if it obtained some new data in previous one
making progress.

This makes it more robust against frequent but transient network failures.
For example:

=> sqlite-src-3370200.zip doesn't seem to exist in /usr/ports/distfiles/.
=> Attempting to fetch https://www.sqlite.org/2022/sqlite-src-3370200.zip
sqlite-src-3370200.zip                          3% of   12 MB   45 kBps 04m24s
fetch: sqlite-src-3370200.zip appears to be truncated: 524288/13145234 bytes
sqlite-src-3370200.zip                         10% of   12 MB   67 kBps 02m56s
fetch: sqlite-src-3370200.zip appears to be truncated: 1327104/13145234 bytes
sqlite-src-3370200.zip                         28% of   12 MB  123 kBps 01m14s
fetch: sqlite-src-3370200.zip appears to be truncated: 3735552/13145234 bytes
sqlite-src-3370200.zip                         54% of   12 MB  253 kBps    24s
fetch: sqlite-src-3370200.zip appears to be truncated: 7176192/13145234 bytes
sqlite-src-3370200.zip                         62% of   12 MB   90 kBps    55s
fetch: sqlite-src-3370200.zip appears to be truncated: 8241152/13145234 bytes
sqlite-src-3370200.zip                         82% of   12 MB  113 kBps    20s
fetch: sqlite-src-3370200.zip appears to be truncated: 10862592/13145234 bytes
sqlite-src-3370200.zip                                  12 MB  185 kBps    12s
===> Fetching all distfiles required by sqlite3-3.37.2,1 for building

MFC after:	1 month

Through a combination of insufficient variable initialization and

2018-09-10T19:39:20+00:00

imprudent reuse of static buffers, the end-of-transfer statistics
displayed when stdout is not a tty always ended up as 0 B / 0 Bps.
Reorganize the code to use caller-provided buffers, tweak the ETA
display a bit, and reduce the visual differences between the tty and
non-tty end-of-transfer displays.

PR:		202424
Approved by:	re (gjb@)