ffprobe: add XML writer

5 files changed, 375 insertions, 2 deletions

diff --git a/Changelog b/Changelog
index 247c97b209..4a0cc3041c 100644
--- a/Changelog
+++ b/Changelog
@@ -9,6 +9,7 @@ version next:
 - SMJPEG demuxer
 - dv: add timecode to metadata
 - thumbnail video filter
+- XML output in ffprobe
 
 
 version 0.9:
diff --git a/Makefile b/Makefile
index 5622aa8a2f..52645a366e 100644
--- a/Makefile
+++ b/Makefile
@@ -39,7 +39,7 @@ FFLIBS-$(CONFIG_SWSCALE)  += swscale
 
 FFLIBS := avutil
 
-DATA_FILES := $(wildcard $(SRC_PATH)/presets/*.ffpreset)
+DATA_FILES := $(wildcard $(SRC_PATH)/presets/*.ffpreset) doc/ffprobe.xsd
 
 SKIPHEADERS = cmdutils_common_opts.h
 
diff --git a/doc/ffprobe.texi b/doc/ffprobe.texi
index 4af13d69e5..736c7f06c0 100644
--- a/doc/ffprobe.texi
+++ b/doc/ffprobe.texi
@@ -218,6 +218,39 @@ Each section is printed using JSON notation.
 
 For more information about JSON, see @url{http://www.json.org/}.
 
+@section xml
+XML based format.
+
+The XML output is described in the XML schema description file
+@file{ffprobe.xsd} installed in the FFmpeg datadir.
+
+Note that the output issued will be compliant to the
+@file{ffprobe.xsd} schema only when no special global output options
+(@option{unit}, @option{prefix}, @option{byte_binary_prefix},
+@option{sexagesimal} etc.) are specified.
+
+This writer accepts options as a list of @var{key}=@var{value} pairs,
+separated by ":".
+
+The description of the accepted options follows.
+
+@table @option
+
+@item fully_qualified, q
+If set to 1 specify if the output should be fully qualified. Default
+value is 0.
+This is required for generating an XML file which can be validated
+through an XSD file.
+
+@item xsd_compliant, x
+If set to 1 perform more checks for ensuring that the output is XSD
+compliant. Default value is 0.
+This option automatically sets @option{fully_qualified} to 1.
+@end table
+
+For more information about the XML format, see
+@url{http://www.w3.org/XML/}.
+
 @c man end WRITERS
 
 @include decoders.texi
diff --git a/doc/ffprobe.xsd b/doc/ffprobe.xsd
new file mode 100644
index 0000000000..92bc2b1b55
--- /dev/null
+++ b/doc/ffprobe.xsd
@@ -0,0 +1,96 @@
+<?xml version="1.0" encoding="UTF-8"?>
+
+<xsd:schema xmlns:xsd="http://www.w3.org/2001/XMLSchema"
+    targetNamespace="http://www.ffmpeg.org/schema/ffprobe"
+    xmlns:ffprobe="http://www.ffmpeg.org/schema/ffprobe">
+
+    <xsd:element name="ffprobe" type="ffprobe:ffprobeType"/>
+
+    <xsd:complexType name="ffprobeType">
+        <xsd:sequence>
+            <xsd:element name="packets"  type="ffprobe:packetsType" minOccurs="0" maxOccurs="1" />
+            <xsd:element name="streams"  type="ffprobe:streamsType" minOccurs="0" maxOccurs="1" />
+            <xsd:element name="format"   type="ffprobe:formatType"  minOccurs="0" maxOccurs="1" />
+        </xsd:sequence>
+    </xsd:complexType>
+
+    <xsd:complexType name="packetsType">
+        <xsd:sequence>
+            <xsd:element name="packet" type="ffprobe:packetType" minOccurs="0" maxOccurs="unbounded"/>
+        </xsd:sequence>
+    </xsd:complexType>
+
+    <xsd:complexType name="packetType">
+      <xsd:attribute name="codec_type"    type="xsd:string" use="required" />
+      <xsd:attribute name="stream_index"  type="xsd:int" use="required" />
+      <xsd:attribute name="pts"           type="xsd:long"  />
+      <xsd:attribute name="pts_time"      type="xsd:float" />
+      <xsd:attribute name="dts"           type="xsd:long"  />
+      <xsd:attribute name="dts_time"      type="xsd:float" />
+      <xsd:attribute name="duration"      type="xsd:long"  />
+      <xsd:attribute name="duration_time" type="xsd:float" />
+      <xsd:attribute name="size"          type="xsd:long" use="required" />
+      <xsd:attribute name="pos"           type="xsd:long"  />
+      <xsd:attribute name="flags"         type="xsd:string" use="required" />
+    </xsd:complexType>
+
+    <xsd:complexType name="streamsType">
+        <xsd:sequence>
+            <xsd:element name="stream" type="ffprobe:streamType" minOccurs="0" maxOccurs="unbounded"/>
+        </xsd:sequence>
+    </xsd:complexType>
+
+    <xsd:complexType name="streamType">
+      <xsd:attribute name="index"            type="xsd:int" use="required"/>
+      <xsd:attribute name="codec_name"       type="xsd:string" />
+      <xsd:attribute name="codec_long_name"  type="xsd:string" />
+      <xsd:attribute name="codec_type"       type="xsd:string" />
+      <xsd:attribute name="codec_time_base"  type="xsd:string" use="required"/>
+      <xsd:attribute name="codec_tag"        type="xsd:string" use="required"/>
+      <xsd:attribute name="codec_tag_string" type="xsd:string" use="required"/>
+
+      <!-- video attributes -->
+      <xsd:attribute name="width"                type="xsd:int"/>
+      <xsd:attribute name="height"               type="xsd:int"/>
+      <xsd:attribute name="has_b_frames"         type="xsd:int"/>
+      <xsd:attribute name="sample_aspect_ratio"  type="xsd:string"/>
+      <xsd:attribute name="display_aspect_ratio" type="xsd:string"/>
+      <xsd:attribute name="pix_fmt"              type="xsd:string"/>
+      <xsd:attribute name="level"                type="xsd:int"/>
+      <xsd:attribute name="timecode"             type="xsd:string"/>
+
+      <!-- audio attributes -->
+      <xsd:attribute name="sample_fmt"       type="xsd:string"/>
+      <xsd:attribute name="sample_rate"      type="xsd:int"/>
+      <xsd:attribute name="channels"         type="xsd:int"/>
+      <xsd:attribute name="bits_per_sample"  type="xsd:int"/>
+
+      <xsd:attribute name="id"               type="xsd:string"/>
+      <xsd:attribute name="r_frame_rate"     type="xsd:string" use="required"/>
+      <xsd:attribute name="avg_frame_rate"   type="xsd:string" use="required"/>
+      <xsd:attribute name="time_base"        type="xsd:string" use="required"/>
+      <xsd:attribute name="start_time"       type="xsd:float"/>
+      <xsd:attribute name="duration"         type="xsd:float"/>
+      <xsd:attribute name="nb_frames"        type="xsd:int"/>
+    </xsd:complexType>
+
+    <xsd:complexType name="formatType">
+      <xsd:sequence>
+        <xsd:element name="tag" type="ffprobe:tagType" minOccurs="0" maxOccurs="unbounded"/>
+      </xsd:sequence>
+
+      <xsd:attribute name="filename"         type="xsd:string" use="required"/>
+      <xsd:attribute name="nb_streams"       type="xsd:int"    use="required"/>
+      <xsd:attribute name="format_name"      type="xsd:string" use="required"/>
+      <xsd:attribute name="format_long_name" type="xsd:string" use="required"/>
+      <xsd:attribute name="start_time"       type="xsd:float"/>
+      <xsd:attribute name="duration"         type="xsd:float"/>
+      <xsd:attribute name="size"             type="xsd:long"/>
+      <xsd:attribute name="bit_rate"         type="xsd:long"/>
+    </xsd:complexType>
+
+    <xsd:complexType name="tagType">
+      <xsd:attribute name="key"   type="xsd:string" use="required"/>
+      <xsd:attribute name="value" type="xsd:string" use="required"/>
+    </xsd:complexType>
+</xsd:schema>
diff --git a/ffprobe.c b/ffprobe.c
index e377c3c767..bf9b580f18 100644
--- a/ffprobe.c
+++ b/ffprobe.c
@@ -867,6 +867,248 @@ static const Writer json_writer = {
     .show_tags            = json_show_tags,
 };
 
+/* XML output */
+
+typedef struct {
+    const AVClass *class;
+    int within_tag;
+    int multiple_entries; ///< tells if the given chapter requires multiple entries
+    int indent_level;
+    int fully_qualified;
+    int xsd_strict;
+    char *buf;
+    size_t buf_size;
+} XMLContext;
+
+#undef OFFSET
+#define OFFSET(x) offsetof(XMLContext, x)
+
+static const AVOption xml_options[] = {
+    {"fully_qualified", "specify if the output should be fully qualified", OFFSET(fully_qualified), AV_OPT_TYPE_INT, {.dbl=0},  0, 1 },
+    {"q",               "specify if the output should be fully qualified", OFFSET(fully_qualified), AV_OPT_TYPE_INT, {.dbl=0},  0, 1 },
+    {"xsd_strict",      "ensure that the output is XSD compliant",         OFFSET(xsd_strict),      AV_OPT_TYPE_INT, {.dbl=0},  0, 1 },
+    {"x",               "ensure that the output is XSD compliant",         OFFSET(xsd_strict),      AV_OPT_TYPE_INT, {.dbl=0},  0, 1 },
+    {NULL},
+};
+
+static const char *xml_get_name(void *ctx)
+{
+    return "xml";
+}
+
+static const AVClass xml_class = {
+    "XMLContext",
+    xml_get_name,
+    xml_options
+};
+
+static av_cold int xml_init(WriterContext *wctx, const char *args, void *opaque)
+{
+    XMLContext *xml = wctx->priv;
+    int err;
+
+    xml->class = &xml_class;
+    av_opt_set_defaults(xml);
+
+    if (args &&
+        (err = (av_set_options_string(xml, args, "=", ":"))) < 0) {
+        av_log(wctx, AV_LOG_ERROR, "Error parsing options string: '%s'\n", args);
+        return err;
+    }
+
+    if (xml->xsd_strict) {
+        xml->fully_qualified = 1;
+#define CHECK_COMPLIANCE(opt, opt_name)                                 \
+        if (opt) {                                                      \
+            av_log(wctx, AV_LOG_ERROR,                                  \
+                   "XSD-compliant output selected but option '%s' was selected, XML output may be non-compliant.\n" \
+                   "You need to disable such option with '-no%s'\n", opt_name, opt_name); \
+        }
+        CHECK_COMPLIANCE(show_private_data, "private");
+        CHECK_COMPLIANCE(show_value_unit,   "unit");
+        CHECK_COMPLIANCE(use_value_prefix,  "prefix");
+    }
+
+    xml->buf_size = ESCAPE_INIT_BUF_SIZE;
+    if (!(xml->buf = av_malloc(xml->buf_size)))
+        return AVERROR(ENOMEM);
+    return 0;
+}
+
+static av_cold void xml_uninit(WriterContext *wctx)
+{
+    XMLContext *xml = wctx->priv;
+    av_freep(&xml->buf);
+}
+
+static const char *xml_escape_str(char **dst, size_t *dst_size, const char *src,
+                                  void *log_ctx)
+{
+    const char *p;
+    char *q;
+    size_t size = 1;
+
+    /* precompute size */
+    for (p = src; *p; p++, size++) {
+        ESCAPE_CHECK_SIZE(src, size, SIZE_MAX-10);
+        switch (*p) {
+        case '&' : size += strlen("&amp;");  break;
+        case '<' : size += strlen("&lt;");   break;
+        case '>' : size += strlen("&gt;");   break;
+        case '\"': size += strlen("&quot;"); break;
+        case '\'': size += strlen("&apos;"); break;
+        default: size++;
+        }
+    }
+    ESCAPE_REALLOC_BUF(dst_size, dst, src, size);
+
+#define COPY_STR(str) {      \
+        const char *s = str; \
+        while (*s)           \
+            *q++ = *s++;     \
+    }
+
+    p = src;
+    q = *dst;
+    while (*p) {
+        switch (*p) {
+        case '&' : COPY_STR("&amp;");  break;
+        case '<' : COPY_STR("&lt;");   break;
+        case '>' : COPY_STR("&gt;");   break;
+        case '\"': COPY_STR("&quot;"); break;
+        case '\'': COPY_STR("&apos;"); break;
+        default: *q++ = *p;
+        }
+        p++;
+    }
+    *q = 0;
+
+    return *dst;
+}
+
+static void xml_print_header(WriterContext *wctx)
+{
+    XMLContext *xml = wctx->priv;
+    const char *qual = " xmlns:xsi='http://www.w3.org/2001/XMLSchema-instance' "
+        "xmlns:ffprobe='http://www.ffmpeg.org/schema/ffprobe' "
+        "xsi:schemaLocation='http://www.ffmpeg.org/schema/ffprobe ffprobe.xsd'";
+
+    printf("<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n");
+    printf("<%sffprobe%s>\n",
+           xml->fully_qualified ? "ffprobe:" : "",
+           xml->fully_qualified ? qual : "");
+
+    xml->indent_level++;
+}
+
+static void xml_print_footer(WriterContext *wctx)
+{
+    XMLContext *xml = wctx->priv;
+
+    xml->indent_level--;
+    printf("</%sffprobe>\n", xml->fully_qualified ? "ffprobe:" : "");
+}
+
+#define XML_INDENT() { int i; for (i = 0; i < xml->indent_level; i++) printf(INDENT); }
+
+static void xml_print_chapter_header(WriterContext *wctx, const char *chapter)
+{
+    XMLContext *xml = wctx->priv;
+
+    if (wctx->nb_chapter)
+        printf("\n");
+    xml->multiple_entries = !strcmp(chapter, "packets") || !strcmp(chapter, "streams");
+
+    if (xml->multiple_entries) {
+        XML_INDENT(); printf("<%s>\n", chapter);
+        xml->indent_level++;
+    }
+}
+
+static void xml_print_chapter_footer(WriterContext *wctx, const char *chapter)
+{
+    XMLContext *xml = wctx->priv;
+
+    if (xml->multiple_entries) {
+        xml->indent_level--;
+        XML_INDENT(); printf("</%s>\n", chapter);
+    }
+}
+
+static void xml_print_section_header(WriterContext *wctx, const char *section)
+{
+    XMLContext *xml = wctx->priv;
+
+    XML_INDENT(); printf("<%s ", section);
+    xml->within_tag = 1;
+}
+
+static void xml_print_section_footer(WriterContext *wctx, const char *section)
+{
+    XMLContext *xml = wctx->priv;
+
+    if (xml->within_tag)
+        printf("/>\n");
+    else {
+        XML_INDENT(); printf("</%s>\n", section);
+    }
+}
+
+static void xml_print_str(WriterContext *wctx, const char *key, const char *value)
+{
+    XMLContext *xml = wctx->priv;
+
+    if (wctx->nb_item)
+        printf(" ");
+    printf("%s=\"%s\"", key, xml_escape_str(&xml->buf, &xml->buf_size, value, wctx));
+}
+
+static void xml_print_int(WriterContext *wctx, const char *key, long long int value)
+{
+    if (wctx->nb_item)
+        printf(" ");
+    printf("%s=\"%lld\"", key, value);
+}
+
+static void xml_show_tags(WriterContext *wctx, AVDictionary *dict)
+{
+    XMLContext *xml = wctx->priv;
+    AVDictionaryEntry *tag = NULL;
+    int is_first = 1;
+
+    xml->indent_level++;
+    while ((tag = av_dict_get(dict, "", tag, AV_DICT_IGNORE_SUFFIX))) {
+        if (is_first) {
+            /* close section tag */
+            printf(">\n");
+            xml->within_tag = 0;
+            is_first = 0;
+        }
+        XML_INDENT();
+        printf("<tag key=\"%s\"",
+               xml_escape_str(&xml->buf, &xml->buf_size, tag->key,   wctx));
+        printf(" value=\"%s\"/>\n",
+               xml_escape_str(&xml->buf, &xml->buf_size, tag->value, wctx));
+    }
+    xml->indent_level--;
+}
+
+static Writer xml_writer = {
+    .name                 = "xml",
+    .priv_size            = sizeof(XMLContext),
+    .init                 = xml_init,
+    .uninit               = xml_uninit,
+    .print_header         = xml_print_header,
+    .print_footer         = xml_print_footer,
+    .print_chapter_header = xml_print_chapter_header,
+    .print_chapter_footer = xml_print_chapter_footer,
+    .print_section_header = xml_print_section_header,
+    .print_section_footer = xml_print_section_footer,
+    .print_integer        = xml_print_int,
+    .print_string         = xml_print_str,
+    .show_tags            = xml_show_tags,
+};
+
 static void writer_register_all(void)
 {
     static int initialized;
@@ -879,6 +1121,7 @@ static void writer_register_all(void)
     writer_register(&compact_writer);
     writer_register(&csv_writer);
     writer_register(&json_writer);
+    writer_register(&xml_writer);
 }
 
 #define print_fmt(k, f, ...) do {              \
@@ -1236,7 +1479,7 @@ static const OptionDef options[] = {
     { "pretty", 0, {(void*)&opt_pretty},
       "prettify the format of displayed values, make it more human readable" },
     { "print_format", OPT_STRING | HAS_ARG, {(void*)&print_format},
-      "set the output printing format (available formats are: default, compact, csv, json)", "format" },
+      "set the output printing format (available formats are: default, compact, csv, json, xml)", "format" },
     { "show_format",  OPT_BOOL, {(void*)&do_show_format} , "show format/container info" },
     { "show_packets", OPT_BOOL, {(void*)&do_show_packets}, "show packets info" },
     { "show_streams", OPT_BOOL, {(void*)&do_show_streams}, "show streams info" },