broken links: also inspect C and H files

7 files changed, 52 insertions, 20 deletions

diff --git a/psutil/_psutil_bsd.c b/psutil/_psutil_bsd.c
index 43189a21..3fa93d4b 100644
--- a/psutil/_psutil_bsd.c
+++ b/psutil/_psutil_bsd.c
@@ -7,7 +7,7 @@
  * Platform-specific module methods for FreeBSD and OpenBSD.
 
  * OpenBSD references:
- * - OpenBSD source code: http://anoncvs.spacehopper.org/openbsd-src/
+ * - OpenBSD source code: https://github.com/openbsd/src
  *
  * OpenBSD / NetBSD: missing APIs compared to FreeBSD implementation:
  * - psutil.net_connections()
@@ -234,13 +234,13 @@ psutil_proc_oneshot_info(PyObject *self, PyObject *args) {
     rss = (long)kp.p_vm_rssize * pagesize;
     #ifdef PSUTIL_OPENBSD
         // VMS, this is how ps determines it on OpenBSD:
-        // http://anoncvs.spacehopper.org/openbsd-src/tree/bin/ps/print.c#n461
-        // vms
+        // https://github.com/openbsd/src/blob/
+        //     588f7f8c69786211f2d16865c552afb91b1c7cba/bin/ps/print.c#L505
         vms = (long)(kp.p_vm_dsize + kp.p_vm_ssize + kp.p_vm_tsize) * pagesize;
     #elif PSUTIL_NETBSD
         // VMS, this is how top determines it on NetBSD:
-        // ftp://ftp.iij.ad.jp/pub/NetBSD/NetBSD-release-6/src/external/bsd/
-        //     top/dist/machine/m_netbsd.c
+        // https://github.com/IIJ-NetBSD/netbsd-src/blob/master/external/
+        //     bsd/top/dist/machine/m_netbsd.c
         vms = (long)kp.p_vm_msize * pagesize;
     #endif
         memtext = (long)kp.p_vm_tsize * pagesize;
diff --git a/psutil/_psutil_sunos.c b/psutil/_psutil_sunos.c
index c205a3ac..422d48c7 100644
--- a/psutil/_psutil_sunos.c
+++ b/psutil/_psutil_sunos.c
@@ -306,7 +306,7 @@ psutil_proc_num_ctx_switches(PyObject *self, PyObject *args) {
  * - 'pr_ioch' is a sum of chars read and written, with no distinction
  * - 'pr_inblk' and 'pr_oublk', which should be the number of bytes
  *    read and written, hardly increase and according to:
- *    http://www.brendangregg.com/Perf/paper_diskubyp1.pdf
+ *    http://www.brendangregg.com/Solaris/paper_diskubyp1.pdf
  *    ...they should be meaningless anyway.
  *
 static PyObject*
@@ -326,7 +326,7 @@ proc_io_counters(PyObject* self, PyObject* args) {
     // *and* written.
     // 'pr_inblk' and 'pr_oublk' should be expressed in blocks of
     // 8KB according to:
-    // http://www.brendangregg.com/Perf/paper_diskubyp1.pdf  (pag. 8)
+    // http://www.brendangregg.com/Solaris/paper_diskubyp1.pdf (pag. 8)
     return Py_BuildValue("kkkk",
                          info.pr_ioch,
                          info.pr_ioch,
diff --git a/psutil/arch/bsd/freebsd.c b/psutil/arch/bsd/freebsd.c
index 11594b9d..a4569001 100644
--- a/psutil/arch/bsd/freebsd.c
+++ b/psutil/arch/bsd/freebsd.c
@@ -324,8 +324,8 @@ psutil_proc_threads(PyObject *self, PyObject *args) {
     // Retrieves all threads used by process returning a list of tuples
     // including thread id, user time and system time.
     // Thanks to Robert N. M. Watson:
-    // http://fxr.googlebit.com/source/usr.bin/procstat/
-    //     procstat_threads.c?v=8-CURRENT
+    // http://code.metager.de/source/xref/freebsd/usr.bin/procstat/
+    //     procstat_threads.c
     long pid;
     int mib[4];
     struct kinfo_proc *kip = NULL;
diff --git a/psutil/arch/bsd/freebsd_socks.c b/psutil/arch/bsd/freebsd_socks.c
index c7a26323..4f36ef11 100644
--- a/psutil/arch/bsd/freebsd_socks.c
+++ b/psutil/arch/bsd/freebsd_socks.c
@@ -202,8 +202,7 @@ psutil_get_pid_from_sock(int sock_hash) {
 
 
 // Reference:
-// https://gitorious.org/freebsd/freebsd/source/
-//     f1d6f4778d2044502209708bc167c05f9aa48615:usr.bin/sockstat/sockstat.c
+// https://github.com/freebsd/freebsd/blob/master/usr.bin/sockstat/sockstat.c
 int psutil_gather_inet(int proto, PyObject *py_retlist) {
     struct xinpgen *xig, *exig;
     struct xinpcb *xip;
diff --git a/psutil/arch/bsd/openbsd.c b/psutil/arch/bsd/openbsd.c
index 8891c461..3b3f4449 100644
--- a/psutil/arch/bsd/openbsd.c
+++ b/psutil/arch/bsd/openbsd.c
@@ -415,7 +415,8 @@ psutil_proc_num_fds(PyObject *self, PyObject *args) {
 PyObject *
 psutil_proc_cwd(PyObject *self, PyObject *args) {
     // Reference:
-    // http://anoncvs.spacehopper.org/openbsd-src/tree/bin/ps/print.c#n179
+    // https://github.com/openbsd/src/blob/
+    //     588f7f8c69786211f2d16865c552afb91b1c7cba/bin/ps/print.c#L191
     long pid;
     struct kinfo_proc kp;
     char path[MAXPATHLEN];
diff --git a/psutil/arch/solaris/v10/ifaddrs.c b/psutil/arch/solaris/v10/ifaddrs.c
index 59529e6a..2eb36a3a 100644
--- a/psutil/arch/solaris/v10/ifaddrs.c
+++ b/psutil/arch/solaris/v10/ifaddrs.c
@@ -1,7 +1,7 @@
 /* Refrences:
  * https://lists.samba.org/archive/samba-technical/2009-February/063079.html
  * http://stackoverflow.com/questions/4139405/#4139811
- * https://code.google.com/p/openpgm/source/browse/trunk/openpgm/pgm/getifaddrs.c
+ * https://github.com/steve-o/openpgm/blob/master/openpgm/pgm/getifaddrs.c
  */
 
 #include <string.h>
diff --git a/scripts/internal/check_broken_links.py b/scripts/internal/check_broken_links.py
index 3d1766b3..0ae2b323 100755
--- a/scripts/internal/check_broken_links.py
+++ b/scripts/internal/check_broken_links.py
@@ -55,7 +55,7 @@ HERE = os.path.abspath(os.path.dirname(__file__))
 REGEX = re.compile(
     r'(?:http|ftp|https)?://'
     r'(?:[a-zA-Z]|[0-9]|[$-_@.&+]|[!*\(\),]|(?:%[0-9a-fA-F][0-9a-fA-F]))+')
-REQUEST_TIMEOUT = 10
+REQUEST_TIMEOUT = 15
 # There are some status codes sent by websites on HEAD request.
 # Like 503 by Microsoft, and 401 by Apple
 # They need to be sent GET request
@@ -87,27 +87,26 @@ def find_urls(s):
     return list(set([sanitize_url(x) for x in matches]))
 
 
-def parse_rst(filename):
+def parse_rst(fname):
     """Look for links in a .rst file."""
-    with open(filename) as f:
+    with open(fname) as f:
         text = f.read()
     urls = find_urls(text)
     # HISTORY file has a lot of dead links.
-    if filename == 'HISTORY.rst' and urls:
+    if fname == 'HISTORY.rst' and urls:
         urls = [
             x for x in urls if
             not x.startswith('https://github.com/giampaolo/psutil/issues')]
     return urls
 
 
-def parse_py(filename):
+def parse_py(fname):
     """Look for links in a .py file."""
-    with open(filename) as f:
+    with open(fname) as f:
         lines = f.readlines()
     urls = set()
     for i, line in enumerate(lines):
         for url in find_urls(line):
-            url = urls[0]
             # comment block
             if line.lstrip().startswith('# '):
                 subidx = i + 1
@@ -122,12 +121,45 @@ def parse_py(filename):
     return list(urls)
 
 
+def parse_c(fname):
+    """Look for links in a .py file."""
+    with open(fname) as f:
+        lines = f.readlines()
+    urls = set()
+    for i, line in enumerate(lines):
+        for url in find_urls(line):
+            # comment block //
+            if line.lstrip().startswith('// '):
+                subidx = i + 1
+                while True:
+                    nextline = lines[subidx].strip()
+                    if re.match('^//     .+', nextline):
+                        url += nextline[2:].strip()
+                    else:
+                        break
+                    subidx += 1
+            # comment block /*
+            elif line.lstrip().startswith('* '):
+                subidx = i + 1
+                while True:
+                    nextline = lines[subidx].strip()
+                    if re.match('^\*     .+', nextline):
+                        url += nextline[1:].strip()
+                    else:
+                        break
+                    subidx += 1
+            urls.add(url)
+    return list(urls)
+
+
 def get_urls(fname):
     """Extracts all URLs available in specified fname."""
     if fname.endswith('.rst'):
         return parse_rst(fname)
     elif fname.endswith('.py'):
         return parse_py(fname)
+    elif fname.endswith('.c') or fname.endswith('.h'):
+        return parse_c(fname)
     else:
         return []