Many hyperlinks are disabled.
Use anonymous login
to enable hyperlinks.
Overview
Comment: | Enhance ext/zonefile/README.md to describe the currently available functionality. |
---|---|
Downloads: | Tarball | ZIP archive |
Timelines: | family | ancestors | descendants | both | zonefile |
Files: | files | file ages | folders |
SHA3-256: |
100137c7f68b2d9a329d6d43e6c0a81b |
User & Date: | dan 2018-02-13 19:01:08.333 |
Context
2018-02-13
| ||
20:08 | Fix handling of maxAutoFrameSize parameter. (check-in: d65e585574 user: dan tags: zonefile) | |
19:01 | Enhance ext/zonefile/README.md to describe the currently available functionality. (check-in: 100137c7f6 user: dan tags: zonefile) | |
18:02 | Remove, for now, the "priority" column from the zonefile_files virtual table. (check-in: 8bf5154bc6 user: dan tags: zonefile) | |
Changes
Changes to ext/zonefile/README.md.
1 |
| > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > | | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 | # The Zonefile Extension ## Functionality ### Creating Zonefile Files To create a new zonefile, first create a database table with the following schema: > CREATE TABLE data( > k INTEGER PRIMARY KEY, > frame INTEGER DEFAULT -1, -- frame number. Automatic if -1 > idx INTEGER DEFAULT -1, -- index of entry within frame. Auto if -1 > v BLOB > ); The table may be created in a persistent or temporary database and may take any name, but must contain the columns above. The table must be populated with a row for each key intended to appear in the new zonefile file. Once the table is populated, a zonefile is created using the following SQL: > SELECT zonefile_write(<file>, <table> [, <parameters>]); where <file> is the name of the file to create on disk, <table> is the name of the database table to read and optional argument <parameters> is a JSON object containing various attributes that influence creation of the zonefile file. Currently the only <parameters> attribute supported is <i>maxAutoFrameSize</i>, which sets the maximum frame size in bytes for automatically generated zonefile frames. For example, to create a zonefile named "test.zonefile" based on the contents of database table "test_input" and with a maximum automatic frame size of 4096 bytes: > SELECT zonefile_write('test.zonefile', 'test_input', > '{"maxAutoFrameSize":4096}' > ); ### Using (Reading) Zonefile Files To create a new zonefile table: > CREATE VIRTUAL TABLE z1 USING zonefile; This creates two virtual tables in the database schema. One read-only table named "z1", with a schema equivalent to: > CREATE TABLE z1( -- this whole table is read-only > k INTEGER PRIMARY KEY, > v BLOB, > fileid INTEGER, > frame INTEGER, > ofst INTEGER, > sz INTEGER > ); And a read-write table named "z1_files" with a schema like: > CREATE TABLE z1_files( > filename TEXT PRIMARY KEY, > ekey BLOB, -- encryption key > fileid INTEGER, -- read-only > header JSON HIDDEN -- read-only > ); Both tables are initially empty. To add a zonefile to the index, insert a row into the "z1_files" table: > INSERT INTO z1_files(filename) VALUES(<filename>); Currently, any value provided for any column other than "filename" is ignored. Files are removed from the index by deleting rows from the z1_files table: > DELETE FROM z1_files WHERE filename = <filename>; Once zonefile files have been added to the index, their contents are visible in table "z1". To retrieve the value associated with a single key from one of the zonefile files in the index: > SELECT v FROM z1 WHERE k = <key>; ## Notes * Contrary to the spec, the implementation uses 32-bit (not 16-bit) frame numbers. So the KeyOffsetPair structure becomes: KeyOffsetPair { uint64 key; |
︙ | ︙ | |||
18 19 20 21 22 23 24 | * Multi-byte integer values are big-endian. * The offsets in the ZoneFileIndex.byteOffsetZoneFrame[] array are relative to the offset in ZoneFileHeader.byteOffsetFrames. This is necessary as we may not know the offset of the start of the frame data until after the ZoneFileIndex structure is compressed. | | | 106 107 108 109 110 111 112 113 114 115 | * Multi-byte integer values are big-endian. * The offsets in the ZoneFileIndex.byteOffsetZoneFrame[] array are relative to the offset in ZoneFileHeader.byteOffsetFrames. This is necessary as we may not know the offset of the start of the frame data until after the ZoneFileIndex structure is compressed. * Currently there is no support at all for encryption or compression. |