cffeed

Description

Reads or creates an RSS or Atom syndication feed. This tag can read RSS versions 0.90, 0.91, 0.92, 0.93, 0.94, 1.0, and 2.0, and Atom 0.3 or 1.0. It can create RSS 2.0 or Atom 1.0 feeds.

Category

Communications tags, Internet protocol tags

Syntax

History

ColdFusion 8: Added this tag.

ColdFusion 9: The escapeChars attribute is newly added.

Attributes

Usage

Specifying file and directory attributes

Use the following syntax to specify an in-memory file or directory. In-memory files are not written to disk and speed processing of transient data.

A filepath can include multiple directories, for example ram:///petStore/images/poodle.jpg. Create the directories in the path before you specify the file. For more information on using in-memory files, see Working with in-memory files in Optimizing ColdFusion applications in the Developing ColdFusion Applications.

Setting and getting feed information

The cffeed tag lets you specify and save feed data in many, flexible ways.

Usage

Use the following syntax to specify an in-memory file, which is not written to disk. In-memory files speed processing of transient data.

The filepath can include directories, for example ram:///petStore/images/poodle.jpg. Create the directories in the path before you specify the file. For more information on using in-memory files, see Working with in-memory files in Optimizing ColdFusion applications in the Developing ColdFusion Applications.

 When you create a feed

• You specify the feed data in either of the following ways:
  • By putting all metadata and entry or item data in a single structure specified by the name attribute.
  • By putting the metadata in a structure specified by the properties structure and the entries or items as rows in a query object specified by the query attribute.
• You save the resulting feed XML in one or both of the following places:
  • A file specified by the OutputFile attribute. The cffeed tag saves the data in UTF-8 encoding.
  • An variable specified by the xmlVar attribute

When you read a feed You can save the feed data in any combination of the following forms:

• By saving all entry or item data and metadata in a single structure specified by the name attribute
• By saving entries or items as rows in a query object specified by the query attribute
• By saving the metadata in a structure specified by the properties structure
• By writing the feed XML in a file specified by the OutputFile attribute. The cffeed tag saves the data in UTF-8 encoding.
• By saving the feed XML in a ColdFusion XML variable specified by the xmlVar attribute
  When you save feed data, you do not have to save both the metadata and the entry or item data. You can specify only the properties attribute, or only the query attribute.

Name and properties structure rules

The name and properties structures must conform to the following rules. For more information on requirements for specific metadata entries, see the section Representing feed metadata.

• All structure key names must be identical to the corresponding feed element names, with the exception of the version and encoding fields. Also, the key names for Dublin Core and Apple iTunes extension elements start with DC_ and ITUNES_ respectively.
• The properties structure fields are identical to the metadata fields in the name structure.
• When you read a feed, the structure contains only those elements and attribute values that exist in the feed. For requirements for the create action, see the section Creating feeds.
• If the feed can have multiple elements of the same type (such as entry, item, or link), the name or propertystructure has a single entry that contains the data for all of the elements. The structure entry has the following format:
  
  • The key is the element name (for example, item)
  • The value is an array of structures

  • Each structure in the array represents one element.
    ColdFusion uses an array even if there is only a single element. If an Atom feed has only one link element, for example, you must specify that element in a nameattribute structure by using the following format:

    structureName.link[1]

    For example, to specify a linkmetadata entry in an Atom 1.0 feed, you could use the following code:

    <cfset meta.link = arrayNew(1)> <cfset meta.link[1] = structNew()> <cfset meta.link[1].href = "http://www.myCo,com">

• If an element can have multiple attributes, or can have at least one attribute and a value, the element is represented as a structure, even if the element specifies only one attribute or only a value.

• If an element has one or more attributes and a value (body), the value is in a field of the element structure named value. For example, the text of the summary element for the third entryin an Atom feed would go in a field whose name has the following format:

  structureName.entry[3].summary.value.

• When the cffeed tag reads a feed, it reports dates as follows:Atom: W3C date format, such as 2006-07-11T18:19:00Z.RSS: in RFC 822 Format, such as Thu, 05 Oct 2006 18:19:00 GMT.
• When the cffeed tag creates a feed, you can use W3C or RFC 822 formats for both feed types. You can also use any standard date or date/time format accepted by ColdFusion.

Query object rules

The query object specified by the query attribute conforms to the following rules:

• The query object format supports multiple feed formats, and many feeds do not include all optional feed attributes or elements. As a result:
  • When you read a feed, the returned query object contains entries for all standard RSS and Atom fields, even for fields that are not supported by the feed type. Any columns that are not used by the feed format, or are not used in that specific feed, contain empty strings or undefined values.
  • When you read a feed, the query object contains all iTunes extension fields if the feed contains any iTunes extension elements, and the query object contains all Dublin Core extension fields if the feed contains any Dublin Core extension elements. Otherwise, the query results do not contain any of the extension fields.
  • When you create a feed, the query that you define requires only those columns that contain data for your feed; you can omit unused columns.
• If a feed entry or item has multiple child elements with the same name, the query column represents the element values as a comma-delimited list. RSS 2.0 items can have multiple category elements. Atom 1.0 entries can have multiple category, author, contributor, and link elements. The Dublin Core extensions allow all multiples of all element types.

• Many entry or item elements that can have multiple instances have multiple attributes, not all of which are required for any particular element instance. If an entry or item has multiple instances of an element, and any of those elements omit attributes, ColdFusion represents the omitted attribute in the lists by a space. In XML, an Atom entry, for example, might contain three author elements, as follows:

  <author> <person>Anthony</person> <uri>http://www.MyCo.com</uri> <email>Tony@MyCo.com</email> </author> <author> <person>Beverly</person> </author> <author> <person>Cathy</person> <email>cathy@MyCo.com</email> </author>

  The ColdFusion query represents these columns as follows:

The following table lists the columns of the standard query object specified by the query attribute. If an RSS feed includes either Dublin Core extensions or iTunes extensions, the query includes additional columns. For information on these fields, see the sections Dublin Core Extensions and Apple iTunes Extensions.

Representing feed metadata

When you create a feed, the name and properties structures can represent all standard metadata for RSS 2 or Atom 1 feeds, in the format described in the Name and properties structure rules section. Similarly, when you read a feed, the structures represent all received metadata. The following rules apply to specific feed metadata fields in the name and properties structures:

• The version field identifies or specifies the feed version in the form format_versionNumber. For the create action, specify atom_1.0 or rss_2.0. When you read an RSS 0.91 feed, the version field value is rss_0.91U, not rss_0.91.
• The feedExtension field identifies whether the feed includes iTunes or Dublin Core extension content. Valid values are itunes and DublinCore. You do not have to specify this field when you create a feed with iTunes extensions; ColdFusion automatically determines that you have specified extension fields. (You cannot create a feed with Dublin Core extensions.)
• For the read action, an encoding field identifies the XML encoding attribute, such as iso-8859-1. Do not specify an encoding field for a create action. Currently, ColdFusion generates all feeds in UTF-8 format and ignores any encoding value that you specify.
• For RSS feeds, the skiphours field contains a comma-delimited list of up to 24 numbers in the range 0-23, specifying hours of the day when aggregators should not read the feed. The hour beginning at midnight is hour zero. Your application can use the field to decide when to read the feed.
• For RSS feeds, the skipdays field contains a comma-delimited list of up to seven day-name values, specifying days of the week when aggregators must not read the feed. The valid names are Monday, Tuesday, Wednesday, Thursday, Friday, Saturday and Sunday. Your application can use the field to decide when to read the feed.

Dublin Core Extensions

Dublin Core extension elements provide additional metadata about the feed or an item. You can use the cffeed tag to read feeds that include elements that conform to the Dublin Core Metadata Element Set specification as metadata (channel elements) or as item elements. For detailed information Dublin Core extension elements, see the Dublin Core Metadata Element Set specification. At the time this topic was written, this specification was available at http://dublincore.org/documents/dces/.ColdFusion support for Dublin Core extensions has the following limitations:

• You cannot create feeds containing these elements.
• You cannot get Dublin Core extension elements that are contained in a top-level (metadata) image element. ColdFusion ignores these elements.
• ColdFusion supports only the Dublin Core Metadata Element Set. It does not support the additional Dublin Core Metadata Initiative elements and element refinements.
  When feed items include the Dublin Core extensions, the query specified by a query attribute includes all of the columns listed in the following table. If the feed does not include any Dublin Core extension elements, the query does not include the columns. With the exception of the DC_SUBJECT_TAXONOMURI and DC_SUBJECT_VALUE columns, each column name (without the DC_ prefix) corresponds directly to a Dublin Core extension element name.

When you get data for a feed that includes Dublin Core elements as a structure, the element names are identical to the query column names listed above, with the exception of the representation of the Dublin Core subject element. The structure format represents the subject element as a dc_subject entry, which consists of an array of structures. The structures in the array have keys with the names value, for the element value, and taxononmyURI, for the taxonomyURI attribute.

Apple iTunes Extensions

You can use the cffeed tag to create or read feeds that contain elements defined in the Apple iTunes RSS podcast specification. For detailed information on iTunes extension format, see the Apple iTunes RSS specification. At the time this topic was written, this specification was available at http://www.apple.com/itunes/store/podcaststechspecs.html.
You can create feeds with only a subset of the iTunes RSS extensions. When you read a feed, ColdFusion ignores all iTunes extension elements that are not in the supported subset. The following table lists the names of structure entries or query column names for the supported elements. (These names consist of the ITUNES_ prefix followed by the iTunes extension element name.) The table also indicates which elements are used in the metadata, which are used in the individual items, and which can be used in both:

You can also use the following channel elements in the name or properties structures.

Creating feeds

When you create a feed, you specify the feed contents in a name structure or in the combination of a query object and a properties structure. The cffeed tag generates the feed XML and saves in to the variable specified by the xmlVar attribute, the file specified by the outputFile attribute, or both. To create an RSS 2.0 feed you must specify the following metadata fields in a name structure or in a properties structure. All other RSS2.0 metadata fields, and all item fields, are optional.

• title
• link
• description

• version (must be "rss_2.0")
  The cffeed tag does not enforce any rules on the Atom feed structure that it creates. You are responsible for ensuring that the feed is valid.
  In most cases, a database table uses column names that differ from the column names you must use to create the feed. Therefore, use the columnmap attribute to map the input query column names to the required column names. The attribute is a structure whose keys are the column names required by the cffeedtag and whose values are the corresponding input query columns.

  Note: Always capitalize the input query column names irrespective of whether the database column names are capitalized or not.

  The following example creates a feed using the cfartgallery data source's orders table. It maps the orders table ORDERDATE column to the query publisheddate column, the ADDRESS column to the content column, and so on. The sample code then displays the generated query XML to show the results.

  <!--- Get the feed data as a query from the orders table. ---> <cfquery name="getOrders" datasource="cfartgallery"> SELECT * FROM orders </cfquery> <!--- Map the orders column names to the feed query column names. ---> <cfset columnMapStruct = StructNew()> <cfset columnMapStruct.publisheddate = "ORDERDATE"> <cfset columnMapStruct.content = "ADDRESS"> <cfset columnMapStruct.title = "CUSTOMERFIRSTNAME"> <cfset columnMapStruct.rsslink = "ORDERID"> <!--- Set the feed metadata. ---> <cfset meta.title = "Art Orders"> <cfset meta.link = "http://feedlink"> <cfset meta.description = "Orders at the art gallery"> <cfset meta.version = "rss_2.0"> <!--- Create the feed. ---> <cffeed action="create" query="#getOrders#" properties="#meta#" columnMap="#columnMapStruct#" xmlvar="rssXML"> <cfdump var="#XMLParse(rssXML)#">

Reading feeds

The cffeed tag does not validate the feeds that it reads. It can read invalid or loosely formatted feeds, but ignores some or all of the invalid content. For example, if you put more than one rights element in the Atom feed (which invalidates the feed), the cffeed tag ignores the elements after the first one, and doesn't generate an error. Dates and times in feeds that are being read must be in W3C or RFC 822 format. ColdFusion can also read iTunes extension dates in the format normally used by the iTunes music store.

Example

The following example creates an RSS feed. Enter fields for the feed title, link, and description elements. Also enter title, link, and description fields for one item. A second item is optional. The application saves the feed in a createRSSOutput.xml file in the feedTest subdirectory of the directory that contains the CFML page.

The following application is a simple feed reader that handles RSS and Atom feeds. It displays the feed title; for each item or entry, it shows the title as a link, and shows the published date and the item or entry contents. To use this example to read the feed created by the first application, enter the URL for the file the application created, for example, http://localhost:8500/cffeed/feedTest/createRSSOutput.xml.