Source-Changes-HG archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]
[src/trunk]: src/usr.bin/ftp * make parsing of ftp:// urls more RFC 1738 comp...
details: https://anonhg.NetBSD.org/src/rev/284bf3d4f3b7
branches: trunk
changeset: 472372:284bf3d4f3b7
user: lukem <lukem%NetBSD.org@localhost>
date: Wed Apr 28 13:35:40 1999 +0000
description:
* make parsing of ftp:// urls more RFC 1738 compliant;
- the path is split on `/', and each directory is CWD-ed into separately.
(from [standards/7484] by Alan Barrett <apb%iafrica.com@localhost>)
- support a trailing `;type=X' suffix, where X is a,i, or d. (d isn't
implemented, but it is recognised)
- the only non-compliant behaviour is that empty directories sections
(e.g `//') aren't run as `CWD ' - as a lot of ftpds don't like that.
Instead, treat this as a no-op.
* don't support globbing for ftp urls, since that's technically not
RFC compliant.
* fix a couple of man-page nits
diffstat:
usr.bin/ftp/fetch.c | 226 +++++++++++++++++++++++++++++++++++++++++++--------
usr.bin/ftp/ftp.1 | 91 +++++++++++++-------
usr.bin/ftp/ftp.c | 6 +-
3 files changed, 252 insertions(+), 71 deletions(-)
diffs (truncated from 528 to 300 lines):
diff -r 295bed306847 -r 284bf3d4f3b7 usr.bin/ftp/fetch.c
--- a/usr.bin/ftp/fetch.c Wed Apr 28 13:27:18 1999 +0000
+++ b/usr.bin/ftp/fetch.c Wed Apr 28 13:35:40 1999 +0000
@@ -1,4 +1,4 @@
-/* $NetBSD: fetch.c,v 1.52 1999/03/22 07:36:40 lukem Exp $ */
+/* $NetBSD: fetch.c,v 1.53 1999/04/28 13:35:40 lukem Exp $ */
/*-
* Copyright (c) 1997, 1998, 1999 The NetBSD Foundation, Inc.
@@ -38,7 +38,7 @@
#include <sys/cdefs.h>
#ifndef lint
-__RCSID("$NetBSD: fetch.c,v 1.52 1999/03/22 07:36:40 lukem Exp $");
+__RCSID("$NetBSD: fetch.c,v 1.53 1999/04/28 13:35:40 lukem Exp $");
#endif /* not lint */
/*
@@ -75,7 +75,8 @@
UNKNOWN_URL_T=-1,
HTTP_URL_T,
FTP_URL_T,
- FILE_URL_T
+ FILE_URL_T,
+ CLASSIC_URL_T
} url_t;
void aborthttp __P((int));
@@ -246,10 +247,21 @@
* malloc(3)ed strings of the relevant section, and port to
* the number given, or ftpport if ftp://, or httpport if http://.
*
- * XXX: this is not totally RFC1738 compliant; path will have the
- * leading `/' unless it's an ftp:// URL; this makes things easier
- * for file:// and http:// URLs. ftp:// URLs have all leading `/'s
- * removed.
+ * XXX: this is not totally RFC 1738 compliant; path will have the
+ * leading `/' unless it's an ftp:// URL, as this makes things easier
+ * for file:// and http:// URLs. ftp:// URLs have the `/' between the
+ * host and the url-path removed, but any additional leading slashes
+ * in the url-path are retained (because they imply that we should
+ * later do "CWD" with a null argument).
+ *
+ * Examples:
+ * input url output path
+ * --------- -----------
+ * "http://host"; NULL
+ * "http://host/"; "/"
+ * "http://host/dir/file"; "/dir/file"
+ * "ftp://host/dir/file"; "dir/file"
+ * "ftp://host//dir/file"; "/dir/file"
*/
static int
parse_url(url, desc, type, user, pass, host, port, path)
@@ -306,9 +318,8 @@
thost = (char *)xmalloc(len + 1);
strncpy(thost, url, len);
thost[len] = '\0';
- if (*type == FTP_URL_T) /* skip all leading /'s for ftp URLs */
- while (*ep && *ep == '/')
- ep++;
+ if (*type == FTP_URL_T) /* skip first / for ftp URLs */
+ ep++;
*path = xstrdup(ep);
}
@@ -344,8 +355,9 @@
if (debug)
fprintf(ttyout,
"parse_url: user `%s' pass `%s' host %s:%d path `%s'\n",
- *user ? *user : "", *pass ? *pass : "", *host ? *host : "",
- ntohs(*port), *path ? *path : "");
+ *user ? *user : "<null>", *pass ? *pass : "<null>",
+ *host ? *host : "<null>", ntohs(*port),
+ *path ? *path : "<null>");
return (0);
}
@@ -1031,15 +1043,15 @@
char portnum[6]; /* large enough for "65535\0" */
char *host, *path, *dir, *file, *user, *pass;
in_port_t port;
- int dirhasglob, filehasglob, oautologin, rval, xargc;
+ int dirhasglob, filehasglob, oautologin, rval, type, xargc;
+ url_t urltype;
host = path = dir = file = user = pass = NULL;
port = 0;
rval = 1;
+ type = TYPE_I;
if (strncasecmp(url, FTP_URL, sizeof(FTP_URL) - 1) == 0) {
- url_t urltype;
-
if ((parse_url(url, "URL", &urltype, &user, &pass,
&host, &port, &path) == -1) ||
(user != NULL && *user == '\0') ||
@@ -1050,8 +1062,31 @@
}
url_decode(user);
url_decode(pass);
- url_decode(path);
+ /*
+ * Note: Don't url_decode(path) here. We need to keep the
+ * distinction between "/" and "%2F" until later.
+ */
+
+ /* check for trailing ';type=[aid]' */
+ cp = strrchr(path, ';');
+ if (cp != NULL) {
+ if (strcasecmp(cp, ";type=a") == 0)
+ type = TYPE_A;
+ else if (strcasecmp(cp, ";type=i") == 0)
+ type = TYPE_I;
+ else if (strcasecmp(cp, ";type=d") == 0) {
+ warnx(
+ "Directory listing via a URL is not supported");
+ goto cleanup_fetch_ftp;
+ } else {
+ warnx("Invalid suffix `%s' in URL `%s'", cp,
+ url);
+ goto cleanup_fetch_ftp;
+ }
+ *cp = 0;
+ }
} else { /* classic style `host:file' */
+ urltype = CLASSIC_URL_T;
host = xstrdup(url);
cp = strchr(host, ':');
if (cp != NULL) {
@@ -1065,8 +1100,34 @@
/* Extract the file and (if present) directory name. */
dir = path;
if (! EMPTYSTRING(dir)) {
+ /*
+ * If we are dealing with classic `host:path' syntax,
+ * then a path of the form `/file' (resulting from
+ * input of the form `host:/file') means that we should
+ * do "CWD /" before retreiving the file. So we set
+ * dir="/" and file="file".
+ *
+ * But if we are dealing with URLs like
+ * `ftp://host/path' then a path of the form `/file'
+ * (resulting from a URL of the form `ftp://host//file')
+ * means that we should do `CWD ' (with an empty
+ * argument) before retrieving the file. So we set
+ * dir="" and file="file".
+ *
+ * If the path does not contain / at all, we set
+ * dir=NULL. (We get a path without any slashes if
+ * we are dealing with classic `host:file' or URL
+ * `ftp://host/file'.)
+ *
+ * In all other cases, we set dir to a string that does
+ * not include the final '/' that separates the dir part
+ * from the file part of the path. (This will be the
+ * empty string if and only if we are dealing with a
+ * path of the form `/file' resulting from an URL of the
+ * form `ftp://host//file'.)
+ */
cp = strrchr(dir, '/');
- if (cp == dir) {
+ if (cp == dir && urltype == CLASSIC_URL_T) {
file = cp + 1;
dir = "/";
} else if (cp != NULL) {
@@ -1077,15 +1138,19 @@
dir = NULL;
}
}
+ if (urltype == FTP_URL_T && file != NULL) {
+ url_decode(file);
+ /* but still don't url_decode(dir) */
+ }
if (debug)
fprintf(ttyout,
"fetch_ftp: user `%s' pass `%s' host %s:%d path `%s' dir `%s' file `%s'\n",
- user ? user : "", pass ? pass : "",
- host ? host : "", ntohs(port), path ? path : "",
- dir ? dir : "", file ? file : "");
+ user ? user : "<null>", pass ? pass : "<null>",
+ host ? host : "<null>", ntohs(port), path ? path : "<null>",
+ dir ? dir : "<null>", file ? file : "<null>");
dirhasglob = filehasglob = 0;
- if (doglob) {
+ if (doglob && urltype == CLASSIC_URL_T) {
if (! EMPTYSTRING(dir) && strpbrk(dir, "*?[]{}") != NULL)
dirhasglob = 1;
if (! EMPTYSTRING(file) && strpbrk(file, "*?[]{}") != NULL)
@@ -1116,18 +1181,102 @@
goto cleanup_fetch_ftp;
}
- /* Always use binary transfers. */
- setbinary(0, NULL);
+ switch (type) {
+ case TYPE_A:
+ setascii(0, NULL);
+ break;
+ case TYPE_I:
+ setbinary(0, NULL);
+ break;
+ default:
+ errx(1, "fetch_ftp: unknown transfer type %d\n", type);
+ }
+
+ /*
+ * Change directories, if necessary.
+ *
+ * Note: don't use EMPTYSTRING(dir) below, because
+ * dir="" means something different from dir=NULL.
+ */
+ if (dir != NULL && !dirhasglob) {
+ char *nextpart;
- /* Change directories, if necessary. */
- if (! EMPTYSTRING(dir) && !dirhasglob) {
- xargv[0] = "cd";
- xargv[1] = dir;
- xargv[2] = NULL;
- dirchange = 0;
- cd(2, xargv);
- if (! dirchange)
- goto cleanup_fetch_ftp;
+ /*
+ * If we are dealing with a classic `host:path' (urltype
+ * is CLASSIC_URL_T) then we have a raw directory
+ * name (not encoded in any way) and we can change
+ * directories in one step.
+ *
+ * If we are dealing with an `ftp://host/path' URL
+ * (urltype is FTP_URL_T), then RFC 1738 says we need to
+ * send a separate CWD command for each unescaped "/"
+ * in the path, and we have to interpret %hex escaping
+ * *after* we find the slashes. It's possible to get
+ * empty components here, (from multiple adjacent
+ * slashes in the path) and RFC 1738 says that we should
+ * still do `CWD ' (with a null argument) in such cases.
+ *
+ * XXX: many ftp servers don't support `CWD ', so we
+ * just skip the empty components.
+ *
+ * Examples:
+ *
+ * host:file dir=NULL, urltype=CLASSIC_URL_T
+ * "RETR file"
+ * ftp://host/file dir=NULL, urltype=FTP_URL_T
+ * "RETR file"
+ * ftp://host//file dir="", urltype=FTP_URL_T
+ * (no-op), "RETR file"
+ * host:/file dir="/", urltype=CLASSIC_URL_T
+ * "CWD /", "RETR file"
+ * ftp://host///file dir="/", urltype=FTP_URL_T
+ * (no-op), (no-op), "RETR file"
+ * ftp://host/%2F/file dir="%2F", urltype=FTP_URL_T
+ * "CWD /", "RETR file"
+ * ftp://host/foo/file dir="foo", urltype=FTP_URL_T
+ * "CWD foo", "RETR file"
+ * ftp://host/foo/bar/file dir="foo/bar"
+ * "CWD foo", "CWD bar", "RETR file"
+ * ftp://host//foo/bar/file dir="/foo/bar"
+ * (no-op), "CWD foo", "CWD bar", "RETR file"
+ * ftp://host/foo//bar/file dir="foo//bar"
+ * "CWD foo", (no-op), "CWD bar", "RETR file"
+ * ftp://host/%2F/foo/bar/file dir="%2F/foo/bar"
+ * "CWD /", "CWD foo", "CWD bar", "RETR file"
+ * ftp://host/%2Ffoo/bar/file dir="%2Ffoo/bar"
+ * "CWD /foo", "CWD bar", "RETR file"
+ * ftp://host/%2Ffoo%2Fbar/file dir="%2Ffoo%2Fbar"
+ * "CWD /foo/bar", "RETR file"
+ * ftp://host/%2Ffoo%2Fbar%2Ffile dir=NULL
+ * "RETR /foo/bar/file"
+ *
+ * Note that we don't need `dir' after this point.
+ */
+ do {
+ if (urltype == FTP_URL_T) {
+ nextpart = strchr(dir, '/');
+ if (nextpart) {
+ *nextpart = '\0';
+ nextpart++;
+ }
+ url_decode(dir);
+ } else
+ nextpart = NULL;
+ if (debug)
+ fprintf(ttyout, "dir `%s', nextpart `%s'\n",
+ dir ? dir : "<null>",
+ nextpart ? nextpart : "<null>");
+ if (*dir != '\0') {
+ xargv[0] = "cd";
+ xargv[1] = dir;
+ xargv[2] = NULL;
+ dirchange = 0;
+ cd(2, xargv);
+ if (! dirchange)
+ goto cleanup_fetch_ftp;
+ }
+ dir = nextpart;
Home |
Main Index |
Thread Index |
Old Index