What's the difference between application and text in media types and when do they use what?
For example there is text/html but on the other hand it's application/json.
Is this a historical thing?
This is described on the MIME types page of the mozilla documentation.
application refers to any kind of binary data while text is theoretically human readable.
type/subtype
The structure of a MIME type is very simple; it consists of a type and
a subtype, two strings, separated by a '/'. No space is allowed. The
type represents the category and can be a discrete or a multipart
type. The subtype is specific to each type.
A MIME type is case-insensitive but traditionally is written all in lower case.
Further down the page you can find a table containing the discrete types:
Type Description
text Represents any document that contains text and is theoretically human readable
image Represents any kind of images. Videos are not included, though animated images (like animated gif) are described with an image type.
audio Represents any kind of audio files
video Represents any kind of video files
application Represents any kind of binary data.
To answer your question about JSON, while you'll frequently come across JSON that is human readable it's primarily used for containing data and isn't necessarily intended to simply be read. Meanwhile a machine can always easily convert the data from JSON into an object (assuming the JSON is correctly formatted).
Related
I have a requirement where I need to parse the transformations defined in the QVT-Operational file.
I need some way to represent the QVT-Operational transformations in a json, xml or any other serialized format.
In model to model transformation performed using operational QVT in eclipse, I am able to generate a trace file in XML format. The trace file provides details on which element in the source model is mapped to which element in the target model but I also require the transformation logic. So is there any way to either convert the QVT-operational file to xml(or any serialized format) or get transformation details in the trace file ?
Interactively there is no support since use of the unstable internal *.qvtox representation is not encouraged.
However programmatically you may save the compiled Resource to a *.qvtox XMI file and load it again later, provided you use a compatible OCL+QVTo release.
See also https://www.eclipse.org/forums/index.php/mv/msg/1109554/1848331/#msg_1848331
There have been several Q&A's on stackoverflow on the maximum length of media types (also known as MIME types), like this one and this one.
However, none of these seem to incorporate that media types can also have parameters. See for example this document
At a fundamental level, you can specify the type of a media file using a simple MIME type, such as video/mp4 or audio/mpeg. However, many media types—especially those that support video tracks—can benefit from the ability to more precisely describe the format of the data within them. For instance, just describing a video in an MPEG-4 file with the MIME type video/mp4 doesn't say anything about what format the actual media within takes.
[...]
video/webm; codecs="vp8, vorbis"
A WebM file containing VP8 video and/or Vorbis audio.
I've skimmed over RFC6838 but I can't seem to find any limit, nor any mention that the subtype length limit includes parameters.
It seems to me now there is no real limit to media type lengths if parameters are appended to them. Am I wrong?
Raw images are produced by cameras, scanners, etc. and typically give you more post-processing flexibility than JPEG images.
If I’m serving this kind of file from my web server, which media type (a.k.a. MIME type) should I use? I’m specifically interested in the CR2 files produced by recent Canon cameras, but answers could also address Adobe’s Digital Negative (DNG) format or raw files from other manufacturers’ cameras. (I didn’t see anything relevant in this list from the IANA.)
According to this XML file from the (defunct?) GNOME RAW Thumbnailer project, the MIME type for raw images in general is image/x-dcraw. Listed by their file extensions, here are the other file types mentioned in that file:
ARW: image/x-sony-arw
CR2: image/x-canon-cr2
CRW: image/x-canon-crw
DCR: image/x-kodak-dcr
DNG: image/x-adobe-dng
ERF: image/x-epson-erf
K25: image/x-kodak-k25
KDC: image/x-kodak-kdc
MRW: image/x-minolta-mrw
NEF: image/x-nikon-nef
ORF: image/x-olympus-orf
PEF: image/x-pentax-pef
RAF: image/x-fuji-raf
RAW: image/x-panasonic-raw
SR2: image/x-sony-sr2
SRF: image/x-sony-srf
X3F: image/x-sigma-x3f
It might be too late for the author of the question, but might be helpful for others struggling with the same problem.
For .CR2 it is image/x-dcraw.
For .DNG it is image/x-adobe-dng
Using the convention of Gnome raw thumbnailer might be an option but MIME is assigned by IANA based on reference standards and/or requests of the vendors, so is certainly a non-uffical way even though these are the MIME the vendors are using. The list of official IANA MIME can be found here. All the other list are actually unofficialial. Also, the Wikipedia list is often not updated.
If the MIME is not present in the list you could extract it using tools such exiftool and looking for MIME Type.
I would also consider to use application/octet-stream which is an official MIME type and is used for unknown types of binary data as explained in RFC 2046, at least the application will know that it is a binary file.
If you don't like this solution or a MIME used by the vendors and need to disambiguate from multiple RAW image files, I would follow the convention used for other proprietary image types in the standard: image/vnd.<manufacturer>.<filename extension> for instance for NEF by Nikon corporation and .mrv for Minolta:
image/vnd.nikon.nef
image/vnd.minolta.mrw
But of course this is a pure speculation and will be not recognized by other application.
I'm wondering if there is a website or a way to get the list of all the Schema.org schemas under a specific category (ex: Place).
I found the complete list on Schema.org - they also provide a JSON-LD file - but there's no way to filter/order this easily.
My goal is to get the list of all the places in a nice array (PHP, JSON, whatever).
Most vocabularies/ontologies are defined using RDF.
The vocabulary Schema.org is no exception. The canonical/machine-readable representation of Schema.org is provided in HTML+RDFa:
http://schema.org/docs/schema_org_rdfa.html
This file always contains the current release. If you are interested in specific versions of Schema.org, snapshots are available.
As a quick overview:
Each Schema.org type is of type rdfs:Class.
Sub-types are specified with the rdfs:subClassOf property.
Each Schema.org property is of type rdf:Property.
Sub-properties are specified with the rdfs:subPropertyOf property.
The domains of a property (i.e., on which types this property is expected) are specified with the schema:domainIncludes property.
The ranges of a property (i.e., which values are expected) are specified with the schema:rangeIncludes property.
RDFa is a RDF serialization which many RDF parsers support (if not, it’s simple to automatically convert it into other RDF serializations, like Turtle, RDF/XML, or JSON-LD). There are many different tools and libraries available for this purpose. (You should use an RDF-aware tool instead of trying to parse the file as HTML.)
There are a lot of file types that are categorized under application even though they contain only plain text. For example, application/javascript. Where can I find a list of all of these miscategorized mime types?
In other words, how can I determine if it is safe to view a file as plain text?