← Commit list

fa6a46473e

Daniel Stenberg
Daniel Stenberg
2026-01-26
SUCCESS Duration: 318s

Subject

tool_cb_hdr: with -J, use the redirect name as a backup

Binary

No binary (build failed)

Architecture & Compiler Flags

Architecture: x86_64-linux-gnu Compiler: gcc (Ubuntu 11.4.0-1ubuntu1~22.04.3) 11.4.0

—

Build Log

(log yok)

Commit Diff GitHub →

docs/TODO.md +0 -19docs/cmdline-opts/remote-header-name.md +3 -0src/tool_cb_hdr.c +67 -22src/tool_sdecls.h +2 -1tests/data/Makefile.am +1 -1tests/data/test1641 +62 -0tests/data/test1642 +61 -0tests/data/test1643 +71 -0
docs/TODO.md+0 -19
@@ -819,25 +819,6 @@ already transferred before the retry.
See [curl issue 1084](https://github.com/curl/curl/issues/1084)
-## consider filename from the redirected URL with `-O` ?
-
-When a user gives a URL and uses `-O`, and curl follows a redirect to a new
-URL, the filename is not extracted and used from the newly redirected-to URL
-even if the new URL may have a much more sensible filename.
-
-This is clearly documented and helps for security since there is no surprise
-to users which filename that might get overwritten, but maybe a new option
-could allow for this or maybe `-J` should imply such a treatment as well as
-`-J` already allows for the server to decide what filename to use so it
-already provides the "may overwrite any file" risk.
-
-This is extra tricky if the original URL has no filename part at all since
-then the current code path does error out with an error message, and we cannot
-*know* already at that point if curl is redirected to a URL that has a
-filename...
-
-See [curl issue 1241](https://github.com/curl/curl/issues/1241)
-
## retry on network is unreachable
The `--retry` option retries transfers on *transient failures*. We later added
docs/cmdline-opts/remote-header-name.md+3 -0
@@ -34,6 +34,9 @@ this option may provide you with rather unexpected filenames.
This feature uses the name from the `filename` field, it does not yet support
the `filename*` field (filenames with explicit character sets).
+Starting in 8.19.0, curl falls back and uses the filename extracted from the
+last redirect header if no `Content-Disposition:` header provides a filename.
+
**WARNING**: Exercise judicious use of this option, especially on Windows. A
rogue server could send you the name of a DLL or other file that could be
loaded automatically by Windows or some third party software.
src/tool_cb_hdr.c+67 -22
@@ -148,30 +148,40 @@ locdone:
/*
* Copies a filename part and returns an ALLOCATED data buffer.
*/
-static char *parse_filename(const char *ptr, size_t len)
+static char *parse_filename(const char *ptr, size_t len, char stop)
{
char *copy;
char *p;
char *q;
- char stop = '\0';
copy = memdup0(ptr, len);
if(!copy)
return NULL;
p = copy;
- if(*p == '\'' || *p == '"') {
- /* store the starting quote */
- stop = *p;
- p++;
- }
- else
- stop = ';';
+ if(stop) {
+ /* a Content-Disposition: header */
+ if(*p == '\'' || *p == '"') {
+ /* store the starting quote */
+ stop = *p;
+ p++;
+ }
- /* scan for the end letter and stop there */
- q = strchr(p, stop);
- if(q)
- *q = '\0';
+ /* scan for the end letter and stop there */
+ q = strchr(p, stop);
+ if(q)
+ *q = '\0';
+ }
+ else {
+ /* this is a Location: header, so we need to trim off any queries and
+ fragments present */
+ q = strchr(p, '?'); /* trim off query, if present */
+ if(q)
+ *q = '\0';
+ q = strchr(p, '#'); /* trim off fragment, if present */
+ if(q)
+ *q = '\0';
+ }
/* if the filename contains a path, only use filename portion */
q = strrchr(p, '/');
@@ -283,12 +293,44 @@ static size_t save_etag(const char *etag_h, const char *endp,
* Content-Disposition header specifying a filename property.
*/
static size_t content_disposition(const char *str, const char *end,
- size_t cb, struct per_transfer *per)
+ size_t cb, struct per_transfer *per,
+ long response) /* response code */
{
struct HdrCbData *hdrcbdata = &per->hdrcbdata;
struct OutStruct *outs = &per->outs;
- if((cb > 20) && checkprefix("Content-disposition:", str)) {
+ if((cb > 9) && checkprefix("Location:", str) && (response/100 == 3)) {
+ /* Get the name off the location header as a temporary measure in case
+ there is no Content-Disposition */
+ const char *p = &str[9];
+ curlx_str_passblanks(&p);
+ if(p < end) { /* as a precaution */
+ char *filename = parse_filename(p, cb - (p - str), 0);
+ if(filename) {
+ if(outs->stream) {
+ /* indication of problem, get out! */
+ curlx_free(filename);
+ return CURL_WRITEFUNC_ERROR;
+ }
+ if(outs->alloc_filename)
+ curlx_free(outs->filename);
+
+ if(per->config->output_dir) {
+ outs->filename = curl_maprintf("%s/%s", per->config->output_dir,
+ filename);
+ curlx_free(filename);
+ if(!outs->filename)
+ return CURL_WRITEFUNC_ERROR;
+ }
+ else
+ outs->filename = filename;
+ outs->alloc_filename = TRUE;
+ outs->is_cd_filename = TRUE; /* set to avoid clobbering existing files
+ by default */
+ }
+ }
+ }
+ else if((cb > 20) && checkprefix("Content-disposition:", str)) {
const char *p = str + 20;
/* look for the 'filename=' parameter (encoded filenames (*=) are not
supported) */
@@ -312,14 +354,17 @@ static size_t content_disposition(const char *str, const char *end,
}
p += 9;
+ curlx_str_passblanks(&p);
len = cb - (size_t)(p - str);
- filename = parse_filename(p, len);
+ filename = parse_filename(p, len, ';');
if(filename) {
if(outs->stream) {
/* indication of problem, get out! */
curlx_free(filename);
return CURL_WRITEFUNC_ERROR;
}
+ if(outs->alloc_filename)
+ curlx_free(outs->filename);
if(per->config->output_dir) {
outs->filename = curl_maprintf("%s/%s", per->config->output_dir,
@@ -440,13 +485,13 @@ size_t tool_header_cb(char *ptr, size_t size, size_t nmemb, void *userdata)
return rc;
}
- /* Parse the content-disposition header. When honor_cd_filename is true
- other headers may be stored until the content-disposition header is
- reached, at which point the saved headers can be written. That means
- the content_disposition() may return an rc when it has saved a
- different header for writing later. */
+ /* Parse the content-disposition and location headers. When
+ honor_cd_filename is true, other headers may be stored until the
+ content-disposition header is reached, at which point the saved headers
+ can be written. That means the content_disposition() may return an rc
+ when it has saved a different header for writing later. */
else if(hdrcbdata->honor_cd_filename) {
- size_t rc = content_disposition(str, end, cb, per);
+ size_t rc = content_disposition(str, end, cb, per, response);
if(rc)
return rc;
}
src/tool_sdecls.h+2 -1
@@ -36,7 +36,8 @@
* dynamically allocated and 'belongs' to this OutStruct, otherwise FALSE.
*
* 'is_cd_filename' member is TRUE when string pointed by 'filename' has been
- * set using a server-specified Content-Disposition filename, otherwise FALSE.
+ * set using a server-specified Content-Disposition or Location filename,
+ * otherwise FALSE.
*
* 'regular_file' member is TRUE when output goes to a regular file, this also
* implies that output is 'seekable' and 'appendable' and also that member
tests/data/Makefile.am+1 -1
@@ -218,7 +218,7 @@ test1620 test1621 test1622 test1623 \
\
test1630 test1631 test1632 test1633 test1634 test1635 test1636 \
\
-test1640 \
+test1640 test1641 test1642 test1643 \
\
test1650 test1651 test1652 test1653 test1654 test1655 test1656 test1657 \
test1658 \
tests/data/test1641+62 -0
@@ -0,0 +1,62 @@
+<?xml version="1.0" encoding="US-ASCII"?>
+<testcase>
+<info>
+<keywords>
+HTTP
+HTTP GET
+-J
+</keywords>
+</info>
+
+<reply>
+<data nocheck="yes">
+HTTP/1.1 301 OK
+Location: %TESTNUMBER0002
+
+</data>
+<data2 nocheck="yes">
+HTTP/1.1 200 OK
+Date: Tue, 09 Nov 2010 14:49:00 GMT
+Server: test-server/fake
+Content-Length: 6
+Connection: close
+Content-Type: text/html
+Content-Disposition: filename=name%TESTNUMBER; charset=funny; option=strange
+
+12345
+</data2>
+</reply>
+
+# Client-side
+<client>
+<server>
+http
+</server>
+<name>
+HTTP GET with -J, a redirect and Content-Disposition in the second response
+</name>
+<command option="no-output,no-include">
+http://%HOSTIP:%HTTPPORT/%TESTNUMBER -J -L -O --output-dir %LOGDIR
+</command>
+</client>
+
+# Verify data after the test has been "shot"
+<verify>
+<protocol crlf="headers">
+GET /%TESTNUMBER HTTP/1.1
+Host: %HOSTIP:%HTTPPORT
+User-Agent: curl/%VERSION
+Accept: */*
+
+GET /%TESTNUMBER0002 HTTP/1.1
+Host: %HOSTIP:%HTTPPORT
+User-Agent: curl/%VERSION
+Accept: */*
+
+</protocol>
+<file name="%LOGDIR/name%TESTNUMBER">
+12345
+</file>
+
+</verify>
+</testcase>
tests/data/test1642+61 -0
@@ -0,0 +1,61 @@
+<?xml version="1.0" encoding="US-ASCII"?>
+<testcase>
+<info>
+<keywords>
+HTTP
+HTTP GET
+-J
+</keywords>
+</info>
+
+<reply>
+<data nocheck="yes">
+HTTP/1.1 301 OK
+Location: go/here/%TESTNUMBER0002
+
+</data>
+<data2 nocheck="yes">
+HTTP/1.1 200 OK
+Date: Tue, 09 Nov 2010 14:49:00 GMT
+Server: test-server/fake
+Content-Length: 6
+Connection: close
+Content-Type: text/html
+
+12345
+</data2>
+</reply>
+
+# Client-side
+<client>
+<server>
+http
+</server>
+<name>
+HTTP GET with -J, redirect, no Content-Disposition use the Location: name
+</name>
+<command option="no-output,no-include">
+http://%HOSTIP:%HTTPPORT/%TESTNUMBER -J -L -O --output-dir %LOGDIR
+</command>
+</client>
+
+# Verify data after the test has been "shot"
+<verify>
+<protocol crlf="headers">
+GET /%TESTNUMBER HTTP/1.1
+Host: %HOSTIP:%HTTPPORT
+User-Agent: curl/%VERSION
+Accept: */*
+
+GET /go/here/%TESTNUMBER0002 HTTP/1.1
+Host: %HOSTIP:%HTTPPORT
+User-Agent: curl/%VERSION
+Accept: */*
+
+</protocol>
+<file name="%LOGDIR/%TESTNUMBER0002">
+12345
+</file>
+
+</verify>
+</testcase>
tests/data/test1643+71 -0
@@ -0,0 +1,71 @@
+<?xml version="1.0" encoding="US-ASCII"?>
+<testcase>
+<info>
+<keywords>
+HTTP
+HTTP GET
+-J
+</keywords>
+</info>
+
+<reply>
+<data nocheck="yes">
+HTTP/1.1 301 OK
+Location: go/here/%TESTNUMBER0002#first
+
+</data>
+<data2 nocheck="yes">
+HTTP/1.1 301 OK
+Location: too/%TESTNUMBER0003?fooo#second/%TESTNUMBER0003
+
+</data2>
+<data3 nocheck="yes">
+HTTP/1.1 200 OK
+Date: Tue, 09 Nov 2010 14:49:00 GMT
+Server: test-server/fake
+Content-Length: 6
+Connection: close
+Content-Type: text/html
+
+12345
+</data3>
+</reply>
+
+# Client-side
+<client>
+<server>
+http
+</server>
+<name>
+HTTP -J, two redirects, use the last Location: name
+</name>
+<command option="no-output,no-include">
+http://%HOSTIP:%HTTPPORT/%TESTNUMBER -J -L -O --output-dir %LOGDIR
+</command>
+</client>
+
+# Verify data after the test has been "shot"
+<verify>
+<protocol crlf="headers">
+GET /%TESTNUMBER HTTP/1.1
+Host: %HOSTIP:%HTTPPORT
+User-Agent: curl/%VERSION
+Accept: */*
+
+GET /go/here/%TESTNUMBER0002 HTTP/1.1
+Host: %HOSTIP:%HTTPPORT
+User-Agent: curl/%VERSION
+Accept: */*
+
+GET /go/here/too/%TESTNUMBER0003?fooo HTTP/1.1
+Host: %HOSTIP:%HTTPPORT
+User-Agent: curl/%VERSION
+Accept: */*
+
+</protocol>
+<file name="%LOGDIR/%TESTNUMBER0003">
+12345
+</file>
+
+</verify>
+</testcase>