You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@nutch.apache.org by Apache Wiki <wi...@apache.org> on 2005/08/09 02:32:15 UTC
[Nutch Wiki] Trivial Update of "NutchFileFormats" by KenKrugler
Dear Wiki user,
You have subscribed to a wiki page or wiki category on "Nutch Wiki" for change notification.
The following page has been changed by KenKrugler:
http://wiki.apache.org/nutch/NutchFileFormats
The comment on the change is:
Wiki formatting cleanup.
------------------------------------------------------------------------------
A segment now consists of five subdirectories, each containing an ArrayFile:
+ {{{#!CSV ,
+ Subdirectory,Value datatype,Variable
- Subdirectory Value datatype Variable
-
- ----+
- ----+
- ----
- fetchlist net.nutch.pagedb.FetchListEntry fetchList
+ fetchlist,net.nutch.pagedb.FetchListEntry,fetchList
+ fetcher,net.nutch.fetcher.FetcherOutput,fetcherWriter
+ content,net.nutch.protocol.Content,contentWriter
+ parse_text,net.nutch.parse.ParseText,parseTextWriter
+ parse_data,net.nutch.parse.ParseData,parseDataWriter
+ }}}
- fetcher net.nutch.fetcher.FetcherOutput fetcherWriter
- content net.nutch.protocol.Content contentWriter
- parse_text net.nutch.parse.ParseText parseTextWriter
- parse_data net.nutch.parse.ParseData parseDataWriter
-
- ----+
- ----+
- ----
FetcherOutput is changed:
+ {{{
1 byte version (value 4, was 3)
FetchListEntry as specified above
16 bytes MD5 hash
1 byte status
8 bytes (long) Java milliseconds fetchdate
+ }}}
New class: net.nutch.protocol.Content
+ {{{
1 byte version (value 1)
UTF8 string url
UTF8 string base
compressed byte array content
UTF8 string contentType
java.util.Properties metadata
+ }}}
New class: net.nutch.parse.ParseText
+ {{{
1 byte version (value 1)
compressed byte array text
+ }}}
New class: net.nutch.parse.ParseData
+ {{{
1 byte version (value 1)
UTF8 string title
4 bytes integer totalOutlinks
@@ -58, +59 @@
UTF8 string URL
UTF8 string anchor
java.util.Properties metadata
+ }}}
== Nutch version 0.4 ==
@@ -71, +73 @@
Nutch relies heavily on mappings (associative arrays) from keys to values. The class net.nutch.io.SequenceFile is a flat file of keys and values. The first four bytes of each such file are ASCII "SEQ" and \001 (C-a), followed by the Java class names of keys and values, written as UTF8 strings, e.g. "SEQ\001\000\004long\000\004long", for a mapping from long integers to long integers. After that follows the key-value pairs. Each pair is introduced by four bytes telling the length in bytes of the pair (excluding the eight length bytes) and four bytes telling the length of the key. The typical long (64 bit) integer is 8 bytes and a long-to-long mapping will have pairs of length 16 bytes, e.g.
+ {{{
00 00 00 10 int length of pair = 0x10 = 16 bytes
00 00 00 08 int length of key = 0x08 = 8 bytes
00 00 00 00 00 00 02 80 long key = 0x280 = 640
00 00 00 00 00 0a 42 9b long value = 0xa429b = 672411
+ }}}
To economize the handling of large data volumes, net.nutch.io.MapFile manages a mapping as two separate files in a subdirectory of its own. The large "data" file stores all keys and values, sorted by the key. The much smaller "index" file points to byte offsets in the data file for a small sample of keys. Only the index file is read into memory.
@@ -84, +88 @@
When Nutch crawls the web, each resulting segment has four subdirectories, each containing an ArrayFile (a MapFile having keys that are long integers):
+ {{{#!CSV ,
+ Subdirectory,Value datatype,Variable
- Subdirectory Value datatype Variable
-
- ----+
- ----+
- ----
- fetchlist net.nutch.pagedb.FetchListEntry fetchList
+ fetchlist,net.nutch.pagedb.FetchListEntry,fetchList
- fetcher net.nutch.fetcher.FetcherOutput fetcherDb
+ fetcher,net.nutch.fetcher.FetcherOutput,fetcherDb
- fetcher_content net.nutch.fetcher.FetcherContent rawDb
+ fetcher_content,net.nutch.fetcher.FetcherContent,rawDb
- fetcher_text net.nutch.fetcher.FetcherText strippedDb
+ fetcher_text,net.nutch.fetcher.FetcherText,strippedDb
+ }}}
-
- ----+
- ----+
- ----
Crawling is performed by net.nutch.fetcher.Fetcher which starts a number of parallel FetcherThread?. Each thread gets an URL from the fetchList, checks robots.txt, retrieves the contents and appends the results to fetcherDb, rawDb, and strippedDb.
The FetchListEntry is represented thus:
+ {{{
1 byte version (value should be 2),
1 byte flag (value 1 = true if page should be fetched)
page, as defined by net.nutch.db.Page:
@@ -115, +114 @@
4 bytes Java float next score
4 bytes number of anchors
a list of anchors represented as UTF8 strings
+ }}}
The FetcherOutput is all of the fetcher's output except the raw and stripped versions of the contents:
+ {{{
1 byte version (value 3)
FetchListEntry as specified above
16 bytes MD5 hash
@@ -128, +129 @@
UTF8 string URL
UTF8 string anchor
8 bytes (long) Java milliseconds fetchdate
+ }}}
The FetcherContent is the raw contents stored in GZIP:
+ {{{
1 byte version (value 1)
compressed byte array
+ }}}
The FetcherText is the text conversion of page's content, stored in GZIP:
+ {{{
1 byte version (value 1)
compressed byte array
+ }}}