sub: add an option to filter subtitles by regex

Works as ad-filter. I had some more plans, for example replacing
matching text with different text, but for now it's dropping matches
only. There's a big warning in the manpage that I might change
semantics. For example, I might turn it into a primitive sed.

In a sane world, you'd probably write a simple script that processes
downloaded subtitles before giving them to mpv, and avoid all this
complexity. But we don't live in a sane world, and the sooner you learn
this, the happier you will be. (But I also want to run this on muxed
subtitles.)

This is pretty straightforward. We use POSIX regexes, which are readily
available without additional pain or dependencies. This also means it's
(apparently) not available on win32 (MinGW). The regex list is because I
hate big monolithic regexes, and this makes it slightly better.

Very superficially tested.

DOCS/man/options.rst

@@ -2571,6 +2571,47 @@ Subtitles
 
     Default: ``no``.
 
+``--sub-filter-regex-...=...``
+    Set a list of regular expressions to match on text subtitles, and remove any
+    lines that match (default: empty). This is a string list option. See
+    `List Options`_ for details. Normally, you should use
+    ``--sub-filter-regex-append=<regex>``, where each option use will append a
+    new regular expression, without having to fight escaping problems.
+
+    List items are matched in order. If a regular expression matches, the
+    process is stopped, and the subtitle line is discarded. The text matched
+    against is, currently, always the ``Text`` field of ASS events (if the
+    subtitle format is different, it is always converted). This may include
+    formatting tags. Matching is case-insensitive, but how this is done depends
+    on the libc, and most likely works in ASCII only. It does not work on
+    bitmap/image subtitles. Unavailable on inferior OSes (requires POSIX regex
+    support).
+
+    .. admonition:: Example
+
+        ``--sub-filter-regex-append=opensubtitles\.org`` filters some ads.
+
+    Technically, using a list for matching is redundant, since you could just
+    use a single combined regular expression. But it helps with diagnosis,
+    ease of use, and temporarily disabling or enabling individual filters.
+
+    .. warning::
+
+        This is experimental. The semantics most likely will change, and if you
+        use this, you should be prepared to update the option later. Ideas
+        include replacing the regexes with a very primitive and small subset of
+        sed, or some method to control case-sensitivity.
+
+``--sub-filter-regex-warn=<yes|no>``
+    Log dropped lines with warning log level, instead of verbose (default: no).
+    Helpful for testing.
+
+``--sub-filter-regex-enable=<yes|no>``
+    Whether to enable regex filtering (default: yes). Note that if no regexes
+    are added to the ``--sub-filter-regex`` list, setting this option to ``yes``
+    has no default. It's meant to easily disable or enable filtering
+    temporarily.
+
 ``--sub-create-cc-track=<yes|no>``
     For every video stream, create a closed captions track (default: no). The
     only purpose is to make the track available for selection at the start of

options/options.c

@@ -201,9 +201,15 @@ const struct m_sub_options mp_sub_filter_opts = {
     .opts = (const struct m_option[]){
         OPT_FLAG("sub-filter-sdh", sub_filter_SDH, 0),
         OPT_FLAG("sub-filter-sdh-harder", sub_filter_SDH_harder, 0),
+        OPT_FLAG("sub-filter-regex-enable", rf_enable, 0),
+        OPT_STRINGLIST("sub-filter-regex", rf_items, 0),
+        OPT_FLAG("sub-filter-regex-warn", rf_warn, 0),
         {0}
     },
     .size = sizeof(OPT_BASE_STRUCT),
+    .defaults = &(OPT_BASE_STRUCT){
+        .rf_enable = 1,
+    },
     .change_flags = UPDATE_SUB_FILT,
 };
 

options/options.h

@@ -103,6 +103,9 @@ struct mp_subtitle_opts {
 struct mp_sub_filter_opts {
     int sub_filter_SDH;
     int sub_filter_SDH_harder;
+    int rf_enable;
+    char **rf_items;
+    int rf_warn;
 };
 
 struct mp_osd_render_opts {

sub/filter_regex.c

@@ -0,0 +1,105 @@
+#include <regex.h>
+#include <sys/types.h>
+
+#include "common/common.h"
+#include "common/msg.h"
+#include "misc/bstr.h"
+#include "options/options.h"
+#include "sd.h"
+
+struct priv {
+    int offset;
+    regex_t *regexes;
+    int num_regexes;
+};
+
+static bool rf_init(struct sd_filter *ft)
+{
+    if (strcmp(ft->codec, "ass") != 0)
+        return false;
+
+    if (!ft->opts->rf_enable)
+        return false;
+
+    struct priv *p = talloc_zero(ft, struct priv);
+    ft->priv = p;
+
+    for (int n = 0; ft->opts->rf_items && ft->opts->rf_items[n]; n++) {
+        char *item = ft->opts->rf_items[n];
+
+        MP_TARRAY_GROW(p, p->regexes, p->num_regexes);
+        regex_t *preg = &p->regexes[p->num_regexes];
+
+        int err = regcomp(preg, item, REG_ICASE | REG_EXTENDED | REG_NOSUB);
+        if (err) {
+            char errbuf[512];
+            regerror(err, preg, errbuf, sizeof(errbuf));
+            MP_ERR(ft, "Regular expression error: '%s'\n", errbuf);
+            continue;
+        }
+
+        p->num_regexes += 1;
+    }
+
+    if (!p->num_regexes)
+        return false;
+
+    char *headers = ft->event_format;
+    while (headers && headers[0]) {
+        p->offset += 1;
+        headers = strchr(headers, ',');
+        if (headers)
+            headers += 1;
+    }
+    p->offset -= 1; // removes Start/End, adds ReadOrder
+
+    return true;
+}
+
+static void rf_uninit(struct sd_filter *ft)
+{
+    struct priv *p = ft->priv;
+
+    for (int n = 0; n < p->num_regexes; n++)
+        regfree(&p->regexes[n]);
+}
+
+static struct demux_packet *rf_filter(struct sd_filter *ft,
+                                      struct demux_packet *pkt)
+{
+    struct priv *p = ft->priv;
+    char *line = bstrto0(NULL, (bstr){(char *)pkt->buffer, pkt->len});
+    bool drop = false;
+
+    char *text = line;
+    for (int n = 0; n < p->offset - 1; n++) {
+        text = strchr(text, ',');
+        if (!text) {
+            MP_WARN(ft, "Malformed event: '%s'\n", line);
+            text = line; // shouldn't happen; random fallback
+            break;
+        }
+        text = text + 1;
+    }
+
+    for (int n = 0; n < p->num_regexes; n++) {
+        int err = regexec(&p->regexes[n], text, 0, NULL, 0);
+        if (err == 0) {
+            int level = ft->opts->rf_warn ? MSGL_WARN : MSGL_V;
+            MP_MSG(ft, level, "Matching regex %d => drop: '%s'\n", n, text);
+            drop = true;
+            break;
+        } else if (err != REG_NOMATCH) {
+            MP_WARN(ft, "Error on regexec() on regex %d.\n", n);
+        }
+    }
+
+    talloc_free(line);
+    return drop ? NULL : pkt;
+}
+
+const struct sd_filter_functions sd_filter_regex = {
+    .init   = rf_init,
+    .uninit = rf_uninit,
+    .filter = rf_filter,
+};

sub/sd.h

@@ -87,5 +87,6 @@ struct sd_filter_functions {
 };
 
 extern const struct sd_filter_functions sd_filter_sdh;
+extern const struct sd_filter_functions sd_filter_regex;
 
 #endif

sub/sd_ass.c

@@ -65,6 +65,9 @@ static void fill_plaintext(struct sd *sd, double pts);
 static const struct sd_filter_functions *const filters[] = {
     // Note: list order defines filter order.
     &sd_filter_sdh,
+#if HAVE_POSIX
+    &sd_filter_regex,
+#endif
     NULL,
 };
 

wscript_build.py

@@ -388,6 +388,7 @@ def build(ctx):
         ( "sub/ass_mp.c",                        "libass"),
         ( "sub/dec_sub.c" ),
         ( "sub/draw_bmp.c" ),
+        ( "sub/filter_regex.c",                  "posix" ),
         ( "sub/filter_sdh.c" ),
         ( "sub/img_convert.c" ),
         ( "sub/lavc_conv.c" ),