diff options
author | Yorhel <git@yorhel.nl> | 2016-11-06 09:21:53 +0100 |
---|---|---|
committer | Yorhel <git@yorhel.nl> | 2016-11-06 09:21:53 +0100 |
commit | 35fab522d6b36c4a151d51a0e79e2650b20d29d6 (patch) | |
tree | 502ff354cbdd659375115292d5b1d2013ac2cd1a /indexer | |
parent | aff68205b0f2a6fabcd4e77ddbc72eb19fcf6cdc (diff) |
Indexer: Support HTTP fetching + misc improvements
Diffstat (limited to 'indexer')
-rw-r--r-- | indexer/Cargo.lock | 289 | ||||
-rw-r--r-- | indexer/Cargo.toml | 1 | ||||
-rw-r--r-- | indexer/src/main.rs | 10 | ||||
-rw-r--r-- | indexer/src/pkg.rs | 69 |
4 files changed, 345 insertions, 24 deletions
diff --git a/indexer/Cargo.lock b/indexer/Cargo.lock index de07942..501d8d3 100644 --- a/indexer/Cargo.lock +++ b/indexer/Cargo.lock @@ -5,6 +5,7 @@ dependencies = [ "clap 2.17.1 (registry+https://github.com/rust-lang/crates.io-index)", "encoding 0.3.0-dev (git+https://github.com/lifthrasiir/rust-encoding)", "env_logger 0.3.5 (registry+https://github.com/rust-lang/crates.io-index)", + "hyper 0.9.11 (registry+https://github.com/rust-lang/crates.io-index)", "lazy_static 0.2.1 (registry+https://github.com/rust-lang/crates.io-index)", "libarchive3-sys 0.1.2 (registry+https://github.com/rust-lang/crates.io-index)", "libc 0.2.17 (registry+https://github.com/rust-lang/crates.io-index)", @@ -58,6 +59,17 @@ dependencies = [ ] [[package]] +name = "cookie" +version = "0.2.5" +source = "registry+https://github.com/rust-lang/crates.io-index" +dependencies = [ + "openssl 0.7.14 (registry+https://github.com/rust-lang/crates.io-index)", + "rustc-serialize 0.3.19 (registry+https://github.com/rust-lang/crates.io-index)", + "time 0.1.35 (registry+https://github.com/rust-lang/crates.io-index)", + "url 1.2.3 (registry+https://github.com/rust-lang/crates.io-index)", +] + +[[package]] name = "encoding" version = "0.3.0-dev" source = "git+https://github.com/lifthrasiir/rust-encoding#61e331b0820311572fa00a06349b0f02511e810c" @@ -135,11 +147,70 @@ version = "0.1.3" source = "registry+https://github.com/rust-lang/crates.io-index" [[package]] +name = "gcc" +version = "0.3.38" +source = "registry+https://github.com/rust-lang/crates.io-index" + +[[package]] +name = "gdi32-sys" +version = "0.2.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +dependencies = [ + "winapi 0.2.8 (registry+https://github.com/rust-lang/crates.io-index)", + "winapi-build 0.1.1 (registry+https://github.com/rust-lang/crates.io-index)", +] + +[[package]] name = "hex" version = "0.2.0" source = "registry+https://github.com/rust-lang/crates.io-index" [[package]] +name = "hpack" +version = "0.2.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +dependencies = [ + "log 0.3.6 (registry+https://github.com/rust-lang/crates.io-index)", +] + +[[package]] +name = "httparse" +version = "1.1.2" +source = "registry+https://github.com/rust-lang/crates.io-index" + +[[package]] +name = "hyper" +version = "0.9.11" +source = "registry+https://github.com/rust-lang/crates.io-index" +dependencies = [ + "cookie 0.2.5 (registry+https://github.com/rust-lang/crates.io-index)", + "httparse 1.1.2 (registry+https://github.com/rust-lang/crates.io-index)", + "language-tags 0.2.2 (registry+https://github.com/rust-lang/crates.io-index)", + "log 0.3.6 (registry+https://github.com/rust-lang/crates.io-index)", + "mime 0.2.2 (registry+https://github.com/rust-lang/crates.io-index)", + "num_cpus 1.1.0 (registry+https://github.com/rust-lang/crates.io-index)", + "openssl 0.7.14 (registry+https://github.com/rust-lang/crates.io-index)", + "openssl-verify 0.1.0 (registry+https://github.com/rust-lang/crates.io-index)", + "rustc-serialize 0.3.19 (registry+https://github.com/rust-lang/crates.io-index)", + "solicit 0.4.4 (registry+https://github.com/rust-lang/crates.io-index)", + "time 0.1.35 (registry+https://github.com/rust-lang/crates.io-index)", + "traitobject 0.0.1 (registry+https://github.com/rust-lang/crates.io-index)", + "typeable 0.1.2 (registry+https://github.com/rust-lang/crates.io-index)", + "unicase 1.4.0 (registry+https://github.com/rust-lang/crates.io-index)", + "url 1.2.3 (registry+https://github.com/rust-lang/crates.io-index)", +] + +[[package]] +name = "idna" +version = "0.1.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +dependencies = [ + "matches 0.1.4 (registry+https://github.com/rust-lang/crates.io-index)", + "unicode-bidi 0.2.3 (registry+https://github.com/rust-lang/crates.io-index)", + "unicode-normalization 0.1.2 (registry+https://github.com/rust-lang/crates.io-index)", +] + +[[package]] name = "kernel32-sys" version = "0.2.2" source = "registry+https://github.com/rust-lang/crates.io-index" @@ -149,6 +220,11 @@ dependencies = [ ] [[package]] +name = "language-tags" +version = "0.2.2" +source = "registry+https://github.com/rust-lang/crates.io-index" + +[[package]] name = "lazy_static" version = "0.2.1" source = "registry+https://github.com/rust-lang/crates.io-index" @@ -168,11 +244,24 @@ version = "0.2.17" source = "registry+https://github.com/rust-lang/crates.io-index" [[package]] +name = "libressl-pnacl-sys" +version = "2.1.6" +source = "registry+https://github.com/rust-lang/crates.io-index" +dependencies = [ + "pnacl-build-helper 1.4.10 (registry+https://github.com/rust-lang/crates.io-index)", +] + +[[package]] name = "log" version = "0.3.6" source = "registry+https://github.com/rust-lang/crates.io-index" [[package]] +name = "matches" +version = "0.1.4" +source = "registry+https://github.com/rust-lang/crates.io-index" + +[[package]] name = "md5" version = "0.2.1" source = "registry+https://github.com/rust-lang/crates.io-index" @@ -186,6 +275,65 @@ dependencies = [ ] [[package]] +name = "mime" +version = "0.2.2" +source = "registry+https://github.com/rust-lang/crates.io-index" +dependencies = [ + "log 0.3.6 (registry+https://github.com/rust-lang/crates.io-index)", +] + +[[package]] +name = "num_cpus" +version = "1.1.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +dependencies = [ + "libc 0.2.17 (registry+https://github.com/rust-lang/crates.io-index)", +] + +[[package]] +name = "openssl" +version = "0.7.14" +source = "registry+https://github.com/rust-lang/crates.io-index" +dependencies = [ + "bitflags 0.7.0 (registry+https://github.com/rust-lang/crates.io-index)", + "gcc 0.3.38 (registry+https://github.com/rust-lang/crates.io-index)", + "lazy_static 0.2.1 (registry+https://github.com/rust-lang/crates.io-index)", + "libc 0.2.17 (registry+https://github.com/rust-lang/crates.io-index)", + "openssl-sys 0.7.17 (registry+https://github.com/rust-lang/crates.io-index)", + "openssl-sys-extras 0.7.14 (registry+https://github.com/rust-lang/crates.io-index)", +] + +[[package]] +name = "openssl-sys" +version = "0.7.17" +source = "registry+https://github.com/rust-lang/crates.io-index" +dependencies = [ + "gdi32-sys 0.2.0 (registry+https://github.com/rust-lang/crates.io-index)", + "libc 0.2.17 (registry+https://github.com/rust-lang/crates.io-index)", + "libressl-pnacl-sys 2.1.6 (registry+https://github.com/rust-lang/crates.io-index)", + "pkg-config 0.3.8 (registry+https://github.com/rust-lang/crates.io-index)", + "user32-sys 0.2.0 (registry+https://github.com/rust-lang/crates.io-index)", +] + +[[package]] +name = "openssl-sys-extras" +version = "0.7.14" +source = "registry+https://github.com/rust-lang/crates.io-index" +dependencies = [ + "gcc 0.3.38 (registry+https://github.com/rust-lang/crates.io-index)", + "libc 0.2.17 (registry+https://github.com/rust-lang/crates.io-index)", + "openssl-sys 0.7.17 (registry+https://github.com/rust-lang/crates.io-index)", +] + +[[package]] +name = "openssl-verify" +version = "0.1.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +dependencies = [ + "openssl 0.7.14 (registry+https://github.com/rust-lang/crates.io-index)", +] + +[[package]] name = "phf" version = "0.7.15" source = "registry+https://github.com/rust-lang/crates.io-index" @@ -204,6 +352,14 @@ version = "0.3.8" source = "registry+https://github.com/rust-lang/crates.io-index" [[package]] +name = "pnacl-build-helper" +version = "1.4.10" +source = "registry+https://github.com/rust-lang/crates.io-index" +dependencies = [ + "tempdir 0.3.5 (registry+https://github.com/rust-lang/crates.io-index)", +] + +[[package]] name = "postgres" version = "0.12.0" source = "registry+https://github.com/rust-lang/crates.io-index" @@ -228,6 +384,14 @@ dependencies = [ ] [[package]] +name = "rand" +version = "0.3.14" +source = "registry+https://github.com/rust-lang/crates.io-index" +dependencies = [ + "libc 0.2.17 (registry+https://github.com/rust-lang/crates.io-index)", +] + +[[package]] name = "regex" version = "0.1.80" source = "registry+https://github.com/rust-lang/crates.io-index" @@ -254,11 +418,46 @@ dependencies = [ ] [[package]] +name = "rustc-serialize" +version = "0.3.19" +source = "registry+https://github.com/rust-lang/crates.io-index" + +[[package]] +name = "rustc_version" +version = "0.1.7" +source = "registry+https://github.com/rust-lang/crates.io-index" +dependencies = [ + "semver 0.1.20 (registry+https://github.com/rust-lang/crates.io-index)", +] + +[[package]] +name = "semver" +version = "0.1.20" +source = "registry+https://github.com/rust-lang/crates.io-index" + +[[package]] +name = "solicit" +version = "0.4.4" +source = "registry+https://github.com/rust-lang/crates.io-index" +dependencies = [ + "hpack 0.2.0 (registry+https://github.com/rust-lang/crates.io-index)", + "log 0.3.6 (registry+https://github.com/rust-lang/crates.io-index)", +] + +[[package]] name = "strsim" version = "0.5.1" source = "registry+https://github.com/rust-lang/crates.io-index" [[package]] +name = "tempdir" +version = "0.3.5" +source = "registry+https://github.com/rust-lang/crates.io-index" +dependencies = [ + "rand 0.3.14 (registry+https://github.com/rust-lang/crates.io-index)", +] + +[[package]] name = "term_size" version = "0.2.1" source = "registry+https://github.com/rust-lang/crates.io-index" @@ -286,6 +485,47 @@ dependencies = [ ] [[package]] +name = "time" +version = "0.1.35" +source = "registry+https://github.com/rust-lang/crates.io-index" +dependencies = [ + "kernel32-sys 0.2.2 (registry+https://github.com/rust-lang/crates.io-index)", + "libc 0.2.17 (registry+https://github.com/rust-lang/crates.io-index)", + "winapi 0.2.8 (registry+https://github.com/rust-lang/crates.io-index)", +] + +[[package]] +name = "traitobject" +version = "0.0.1" +source = "registry+https://github.com/rust-lang/crates.io-index" + +[[package]] +name = "typeable" +version = "0.1.2" +source = "registry+https://github.com/rust-lang/crates.io-index" + +[[package]] +name = "unicase" +version = "1.4.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +dependencies = [ + "rustc_version 0.1.7 (registry+https://github.com/rust-lang/crates.io-index)", +] + +[[package]] +name = "unicode-bidi" +version = "0.2.3" +source = "registry+https://github.com/rust-lang/crates.io-index" +dependencies = [ + "matches 0.1.4 (registry+https://github.com/rust-lang/crates.io-index)", +] + +[[package]] +name = "unicode-normalization" +version = "0.1.2" +source = "registry+https://github.com/rust-lang/crates.io-index" + +[[package]] name = "unicode-segmentation" version = "0.1.2" source = "registry+https://github.com/rust-lang/crates.io-index" @@ -301,6 +541,24 @@ version = "0.3.1" source = "registry+https://github.com/rust-lang/crates.io-index" [[package]] +name = "url" +version = "1.2.3" +source = "registry+https://github.com/rust-lang/crates.io-index" +dependencies = [ + "idna 0.1.0 (registry+https://github.com/rust-lang/crates.io-index)", + "matches 0.1.4 (registry+https://github.com/rust-lang/crates.io-index)", +] + +[[package]] +name = "user32-sys" +version = "0.2.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +dependencies = [ + "winapi 0.2.8 (registry+https://github.com/rust-lang/crates.io-index)", + "winapi-build 0.1.1 (registry+https://github.com/rust-lang/crates.io-index)", +] + +[[package]] name = "utf8-ranges" version = "0.1.3" source = "registry+https://github.com/rust-lang/crates.io-index" @@ -327,6 +585,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index" "checksum bufstream 0.1.2 (registry+https://github.com/rust-lang/crates.io-index)" = "7b48dbe2ff0e98fa2f03377d204a9637d3c9816cd431bfe05a8abbd0ea11d074" "checksum byteorder 0.5.3 (registry+https://github.com/rust-lang/crates.io-index)" = "0fc10e8cc6b2580fda3f36eb6dc5316657f812a3df879a44a66fc9f0fdbc4855" "checksum clap 2.17.1 (registry+https://github.com/rust-lang/crates.io-index)" = "27dac76762fb56019b04aed3ccb43a770a18f80f9c2eb62ee1a18d9fb4ea2430" +"checksum cookie 0.2.5 (registry+https://github.com/rust-lang/crates.io-index)" = "0e3d6405328b6edb412158b3b7710e2634e23f3614b9bb1c412df7952489a626" "checksum encoding 0.3.0-dev (git+https://github.com/lifthrasiir/rust-encoding)" = "<none>" "checksum encoding-index-japanese 1.20141219.6 (git+https://github.com/lifthrasiir/rust-encoding)" = "<none>" "checksum encoding-index-korean 1.20141219.6 (git+https://github.com/lifthrasiir/rust-encoding)" = "<none>" @@ -337,29 +596,59 @@ source = "registry+https://github.com/rust-lang/crates.io-index" "checksum encoding_index_tests 0.1.5 (git+https://github.com/lifthrasiir/rust-encoding)" = "<none>" "checksum env_logger 0.3.5 (registry+https://github.com/rust-lang/crates.io-index)" = "15abd780e45b3ea4f76b4e9a26ff4843258dd8a3eed2775a0e7368c2e7936c2f" "checksum fallible-iterator 0.1.3 (registry+https://github.com/rust-lang/crates.io-index)" = "5d48ab1bc11a086628e8cc0cc2c2dc200b884ac05c4b48fb71d6036b6999ff1d" +"checksum gcc 0.3.38 (registry+https://github.com/rust-lang/crates.io-index)" = "553f11439bdefe755bf366b264820f1da70f3aaf3924e594b886beb9c831bcf5" +"checksum gdi32-sys 0.2.0 (registry+https://github.com/rust-lang/crates.io-index)" = "0912515a8ff24ba900422ecda800b52f4016a56251922d397c576bf92c690518" "checksum hex 0.2.0 (registry+https://github.com/rust-lang/crates.io-index)" = "d6a22814455d41612f41161581c2883c0c6a1c41852729b17d5ed88f01e153aa" +"checksum hpack 0.2.0 (registry+https://github.com/rust-lang/crates.io-index)" = "3d2da7d3a34cf6406d9d700111b8eafafe9a251de41ae71d8052748259343b58" +"checksum httparse 1.1.2 (registry+https://github.com/rust-lang/crates.io-index)" = "46534074dbb80b070d60a5cb8ecadd8963a00a438ae1a95268850a7ef73b67ae" +"checksum hyper 0.9.11 (registry+https://github.com/rust-lang/crates.io-index)" = "edd47c66782933e546a32ae89ca3c49263b2ba9bc29f3a0d5c52fff48e0ac67c" +"checksum idna 0.1.0 (registry+https://github.com/rust-lang/crates.io-index)" = "1053236e00ce4f668aeca4a769a09b3bf5a682d802abd6f3cb39374f6b162c11" "checksum kernel32-sys 0.2.2 (registry+https://github.com/rust-lang/crates.io-index)" = "7507624b29483431c0ba2d82aece8ca6cdba9382bff4ddd0f7490560c056098d" +"checksum language-tags 0.2.2 (registry+https://github.com/rust-lang/crates.io-index)" = "a91d884b6667cd606bb5a69aa0c99ba811a115fc68915e7056ec08a46e93199a" "checksum lazy_static 0.2.1 (registry+https://github.com/rust-lang/crates.io-index)" = "49247ec2a285bb3dcb23cbd9c35193c025e7251bfce77c1d5da97e6362dffe7f" "checksum libarchive3-sys 0.1.2 (registry+https://github.com/rust-lang/crates.io-index)" = "3cd3beae8f59a4c7a806523269b5392037577c150446e88d684dfa6de6031ca7" "checksum libc 0.2.17 (registry+https://github.com/rust-lang/crates.io-index)" = "044d1360593a78f5c8e5e710beccdc24ab71d1f01bc19a29bcacdba22e8475d8" +"checksum libressl-pnacl-sys 2.1.6 (registry+https://github.com/rust-lang/crates.io-index)" = "cbc058951ab6a3ef35ca16462d7642c4867e6403520811f28537a4e2f2db3e71" "checksum log 0.3.6 (registry+https://github.com/rust-lang/crates.io-index)" = "ab83497bf8bf4ed2a74259c1c802351fcd67a65baa86394b6ba73c36f4838054" +"checksum matches 0.1.4 (registry+https://github.com/rust-lang/crates.io-index)" = "efd7622e3022e1a6eaa602c4cea8912254e5582c9c692e9167714182244801b1" "checksum md5 0.2.1 (registry+https://github.com/rust-lang/crates.io-index)" = "7df230903ccdffd6b3b4ec21624498ea64c912ce50297846907f0b8e1bb249dd" "checksum memchr 0.1.11 (registry+https://github.com/rust-lang/crates.io-index)" = "d8b629fb514376c675b98c1421e80b151d3817ac42d7c667717d282761418d20" +"checksum mime 0.2.2 (registry+https://github.com/rust-lang/crates.io-index)" = "b5c93a4bd787ddc6e7833c519b73a50883deb5863d76d9b71eb8216fb7f94e66" +"checksum num_cpus 1.1.0 (registry+https://github.com/rust-lang/crates.io-index)" = "8890e6084723d57d0df8d2720b0d60c6ee67d6c93e7169630e4371e88765dcad" +"checksum openssl 0.7.14 (registry+https://github.com/rust-lang/crates.io-index)" = "c4117b6244aac42ed0150a6019b4d953d28247c5dd6ae6f46ae469b5f2318733" +"checksum openssl-sys 0.7.17 (registry+https://github.com/rust-lang/crates.io-index)" = "89c47ee94c352eea9ddaf8e364be7f978a3bb6d66d73176572484238dd5a5c3f" +"checksum openssl-sys-extras 0.7.14 (registry+https://github.com/rust-lang/crates.io-index)" = "11c5e1dba7d3d03d80f045bf0d60111dc69213b67651e7c889527a3badabb9fa" +"checksum openssl-verify 0.1.0 (registry+https://github.com/rust-lang/crates.io-index)" = "3ed86cce894f6b0ed4572e21eb34026f1dc8869cb9ee3869029131bc8c3feb2d" "checksum phf 0.7.15 (registry+https://github.com/rust-lang/crates.io-index)" = "17896951e179a6cbed7d3519b3078ac6c03a347d3e9cf8f303c8a1a73c5a3e44" "checksum phf_shared 0.7.15 (registry+https://github.com/rust-lang/crates.io-index)" = "bb6c14aac1140c2b06b41477096f249416b17c893d56386a892ac657edfdffba" "checksum pkg-config 0.3.8 (registry+https://github.com/rust-lang/crates.io-index)" = "8cee804ecc7eaf201a4a207241472cc870e825206f6c031e3ee2a72fa425f2fa" +"checksum pnacl-build-helper 1.4.10 (registry+https://github.com/rust-lang/crates.io-index)" = "61c9231d31aea845007443d62fcbb58bb6949ab9c18081ee1e09920e0cf1118b" "checksum postgres 0.12.0 (registry+https://github.com/rust-lang/crates.io-index)" = "a7ef92468927003a037e175b54320319e358886865899b37f7318837a646a9fd" "checksum postgres-protocol 0.1.0 (registry+https://github.com/rust-lang/crates.io-index)" = "7e2fc3d800dacc2dd749b690ad15b9b78bc04c26c3f0525cbe163436559bc3fc" +"checksum rand 0.3.14 (registry+https://github.com/rust-lang/crates.io-index)" = "2791d88c6defac799c3f20d74f094ca33b9332612d9aef9078519c82e4fe04a5" "checksum regex 0.1.80 (registry+https://github.com/rust-lang/crates.io-index)" = "4fd4ace6a8cf7860714a2c2280d6c1f7e6a413486c13298bbc86fd3da019402f" "checksum regex-syntax 0.3.9 (registry+https://github.com/rust-lang/crates.io-index)" = "f9ec002c35e86791825ed294b50008eea9ddfc8def4420124fbc6b08db834957" "checksum ring 0.5.3 (registry+https://github.com/rust-lang/crates.io-index)" = "c0743ef007bcff4909b107907a410418eb7e5c6ad55b843d70b39f62bfb7112e" +"checksum rustc-serialize 0.3.19 (registry+https://github.com/rust-lang/crates.io-index)" = "6159e4e6e559c81bd706afe9c8fd68f547d3e851ce12e76b1de7914bab61691b" +"checksum rustc_version 0.1.7 (registry+https://github.com/rust-lang/crates.io-index)" = "c5f5376ea5e30ce23c03eb77cbe4962b988deead10910c372b226388b594c084" +"checksum semver 0.1.20 (registry+https://github.com/rust-lang/crates.io-index)" = "d4f410fedcf71af0345d7607d246e7ad15faaadd49d240ee3b24e5dc21a820ac" +"checksum solicit 0.4.4 (registry+https://github.com/rust-lang/crates.io-index)" = "172382bac9424588d7840732b250faeeef88942e37b6e35317dce98cafdd75b2" "checksum strsim 0.5.1 (registry+https://github.com/rust-lang/crates.io-index)" = "50c069df92e4b01425a8bf3576d5d417943a6a7272fbabaf5bd80b1aaa76442e" +"checksum tempdir 0.3.5 (registry+https://github.com/rust-lang/crates.io-index)" = "87974a6f5c1dfb344d733055601650059a3363de2a6104819293baff662132d6" "checksum term_size 0.2.1 (registry+https://github.com/rust-lang/crates.io-index)" = "3f7f5f3f71b0040cecc71af239414c23fd3c73570f5ff54cf50e03cef637f2a0" "checksum thread-id 2.0.0 (registry+https://github.com/rust-lang/crates.io-index)" = "a9539db560102d1cef46b8b78ce737ff0bb64e7e18d35b2a5688f7d097d0ff03" "checksum thread_local 0.2.7 (registry+https://github.com/rust-lang/crates.io-index)" = "8576dbbfcaef9641452d5cf0df9b0e7eeab7694956dd33bb61515fb8f18cfdd5" +"checksum time 0.1.35 (registry+https://github.com/rust-lang/crates.io-index)" = "3c7ec6d62a20df54e07ab3b78b9a3932972f4b7981de295563686849eb3989af" +"checksum traitobject 0.0.1 (registry+https://github.com/rust-lang/crates.io-index)" = "07eaeb7689bb7fca7ce15628319635758eda769fed481ecfe6686ddef2600616" +"checksum typeable 0.1.2 (registry+https://github.com/rust-lang/crates.io-index)" = "1410f6f91f21d1612654e7cc69193b0334f909dcf2c790c4826254fbb86f8887" +"checksum unicase 1.4.0 (registry+https://github.com/rust-lang/crates.io-index)" = "13a5906ca2b98c799f4b1ab4557b76367ebd6ae5ef14930ec841c74aed5f3764" +"checksum unicode-bidi 0.2.3 (registry+https://github.com/rust-lang/crates.io-index)" = "c1f7ceb96afdfeedee42bade65a0d585a6a0106f681b6749c8ff4daa8df30b3f" +"checksum unicode-normalization 0.1.2 (registry+https://github.com/rust-lang/crates.io-index)" = "26643a2f83bac55f1976fb716c10234485f9202dcd65cfbdf9da49867b271172" "checksum unicode-segmentation 0.1.2 (registry+https://github.com/rust-lang/crates.io-index)" = "b905d0fc2a1f0befd86b0e72e31d1787944efef9d38b9358a9e92a69757f7e3b" "checksum unicode-width 0.1.3 (registry+https://github.com/rust-lang/crates.io-index)" = "2d6722facc10989f63ee0e20a83cd4e1714a9ae11529403ac7e0afd069abc39e" "checksum untrusted 0.3.1 (registry+https://github.com/rust-lang/crates.io-index)" = "5d9bc0e6e73a10975d1fbff8ac3541e221181b0d8998351600fb5523de634c0d" +"checksum url 1.2.3 (registry+https://github.com/rust-lang/crates.io-index)" = "48ccf7bd87a81b769cf84ad556e034541fb90e1cd6d4bc375c822ed9500cd9d7" +"checksum user32-sys 0.2.0 (registry+https://github.com/rust-lang/crates.io-index)" = "4ef4711d107b21b410a3a974b1204d9accc8b10dad75d8324b5d755de1617d47" "checksum utf8-ranges 0.1.3 (registry+https://github.com/rust-lang/crates.io-index)" = "a1ca13c08c41c9c3e04224ed9ff80461d97e121589ff27c753a16cb10830ae0f" "checksum vec_map 0.6.0 (registry+https://github.com/rust-lang/crates.io-index)" = "cac5efe5cb0fa14ec2f84f83c701c562ee63f6dcc680861b21d65c682adfb05f" "checksum winapi 0.2.8 (registry+https://github.com/rust-lang/crates.io-index)" = "167dc9d6949a9b857f3451275e911c3f44255842c1f7a76f33c55103a909087a" diff --git a/indexer/Cargo.toml b/indexer/Cargo.toml index 5c828ba..d8ce4a0 100644 --- a/indexer/Cargo.toml +++ b/indexer/Cargo.toml @@ -14,3 +14,4 @@ encoding = { git = "https://github.com/lifthrasiir/rust-encoding", features = [" ring = "0.5.3" postgres = "0.12.0" clap = "2.16.3" +hyper = "0.9.11" diff --git a/indexer/src/main.rs b/indexer/src/main.rs index 5649528..49c7079 100644 --- a/indexer/src/main.rs +++ b/indexer/src/main.rs @@ -8,6 +8,7 @@ extern crate libc; extern crate ring; extern crate encoding; extern crate postgres; +extern crate hyper; mod archive; mod archread; @@ -29,9 +30,9 @@ fn main() { let arg = clap_app!(indexer => (about: "Manned.org man page indexer") (@arg v: -v +multiple "Increase verbosity") - (@arg host: -h +required +takes_value "PostgreSQL connection string") (@subcommand pkg => (about: "Index a single package") + (@arg force: --force "Overwrite existing indexed package") (@arg sys: --sys +required +takes_value "System short-name") (@arg cat: --cat +required +takes_value "Package category") (@arg pkg: --pkg +required +takes_value "Package name") @@ -52,7 +53,11 @@ fn main() { .filter(Some("postgres"), if verbose >= 4 { log::LogLevelFilter::Trace } else { log::LogLevelFilter::Info }) .init().unwrap(); - let db = match postgres::Connection::connect(arg.value_of("host").unwrap(), postgres::TlsMode::None) { + let dbhost = match std::env::var("MANNED_PG") { + Ok(x) => x, + Err(_) => { error!("MANNED_PG not set."); return } + }; + let db = match postgres::Connection::connect(&dbhost[..], postgres::TlsMode::None) { Ok(x) => x, Err(x) => { error!("Can't connect to postgres: {}", x); return }, }; @@ -60,6 +65,7 @@ fn main() { if let Some(matches) = arg.subcommand_matches("pkg") { pkg::pkg(&db, pkg::PkgOpt { + force: matches.is_present("force"), sys: sysbyshort(&db, matches.value_of("sys").unwrap()), cat: matches.value_of("cat").unwrap(), pkg: matches.value_of("pkg").unwrap(), diff --git a/indexer/src/pkg.rs b/indexer/src/pkg.rs index a2c0c73..4120988 100644 --- a/indexer/src/pkg.rs +++ b/indexer/src/pkg.rs @@ -1,6 +1,7 @@ use std; use std::io::Read; use postgres; +use hyper; use archive; use archread; @@ -8,11 +9,12 @@ use man; use archive::Archive; pub struct PkgOpt<'a> { + pub force: bool, pub sys: i32, pub cat: &'a str, pub pkg: &'a str, pub ver: &'a str, - pub date: &'a str, + pub date: &'a str, // TODO: Option to extract date from package metadata itself pub file: &'a str } @@ -30,17 +32,26 @@ fn insert_pkg(tr: &postgres::transaction::Transaction, opt: &PkgOpt) -> Option<i Ok(r) => r.get(0).get(0), }; - // TODO: option to overwrite an existing package version - let q = "INSERT INTO package_versions (package, version, released) VALUES($1, $2, $3::text::date) RETURNING id"; - let verid: i32 = match tr.query(q, &[&pkgid, &opt.ver, &opt.date]) { - Err(e) => { - error!("Can't insert package version in database: {}", e); - return None; - }, - Ok(r) => r.get(0).get(0), - }; - trace!("Package pkgid {} verid {}", pkgid, verid); - Some(verid) + let q = "SELECT id FROM package_versions WHERE package = $1 AND version = $2 AND released = $3::text::date"; + let res = tr.query(q, &[&pkgid, &opt.ver, &opt.date]).unwrap(); + + let verid : i32; + if res.is_empty() { + let q = "INSERT INTO package_versions (package, version, released) VALUES($1, $2, $3::text::date) RETURNING id"; + verid = tr.query(q, &[&pkgid, &opt.ver, &opt.date]).unwrap().get(0).get(0); + trace!("New package pkgid {} verid {}", pkgid, verid); + Some(verid) + + } else if opt.force { + verid = res.get(0).get(0); + trace!("Overwriting package pkgid {} verid {}", pkgid, verid); + tr.query("DELETE FROM man WHERE package = $1", &[&verid]).unwrap(); + Some(verid) + + } else { + info!("Package already in database, pkgid {} verid {}", pkgid, res.get(0).get::<usize,i32>(0)); + None + } } @@ -50,7 +61,7 @@ fn insert_man_row(tr: &postgres::GenericConnection, verid: i32, path: &str, hash if let Err(e) = tr.execute( "INSERT INTO man (package, name, filename, locale, hash, section) VALUES ($1, $2, '/'||$3, $4, $5, $6)", &[&verid, &name, &path, &locale, &hash, §] - ) { + ) { // I think this can only happen if archread gives us the same file twice, which really // shouldn't happen. But I'd rather continue with an error logged than panic. error!("Can't insert verid {} fn {}: {}", verid, path, e); @@ -64,11 +75,12 @@ fn insert_man(tr: &postgres::GenericConnection, verid: i32, paths: &[&str], ent: Ok(x) => x, }; - // TODO: Overwrite entry if the contents are different? It's possible that earlier decoding + // Overwrite entry if the contents are different. It's possible that earlier decoding // implementations didn't properly detect the encoding. (On the other hand, due to differences - // in filenames it's also possible that THIS decoding step went wrong. Ugh) + // in filenames it's also possible that THIS decoding step went wrong, but that's slightly less + // likely) tr.execute( - "INSERT INTO contents (hash, content) VALUES($1, $2) ON CONFLICT (hash) DO NOTHING", + "INSERT INTO contents (hash, content) VALUES($1, $2) ON CONFLICT (hash) DO UPDATE SET content = $2", &[&dig.as_ref(), &cont] ).unwrap(); @@ -94,11 +106,24 @@ fn insert_link(tr: &postgres::GenericConnection, verid: i32, src: &str, dest: &s fn with_pkg<T,F>(file: &str, cb: F) -> std::io::Result<T> where F: FnOnce(Option<archive::ArchiveEntry>) -> std::io::Result<T> { - // TODO: Support streaming from URLs - // TODO: How does .deb support fit into this? (Or anything else with metadata) - let mut f = try!(std::fs::File::open(file)); - let ent = try!(Archive::open_archive(&mut f)); - cb(ent) + // TODO: .deb support + + if file.starts_with("http://") || file.starts_with("https://") { + let mut res = try!( + hyper::Client::new().get(file).send() + .map_err(|e| std::io::Error::new(std::io::ErrorKind::Other, format!("Hyper: {}", e))) + ); + if !res.status.is_success() { + return Err(std::io::Error::new(std::io::ErrorKind::Other, format!("HTTP: {}", res.status) )); + } + let ent = try!(Archive::open_archive(&mut res)); + cb(ent) + + } else { + let mut res = try!(std::fs::File::open(file)); + let ent = try!(Archive::open_archive(&mut res)); + cb(ent) + } } @@ -121,7 +146,7 @@ fn index_pkg(tr: &postgres::GenericConnection, opt: &PkgOpt, verid: i32) -> std: pub fn pkg(conn: &postgres::GenericConnection, opt: PkgOpt) { - info!("Handling pkg: {} / {} / {} - {} @ {} in {}", opt.sys, opt.cat, opt.pkg, opt.ver, opt.date, opt.file); + info!("Handling pkg: {} / {} / {} - {} @ {} @ {}", opt.sys, opt.cat, opt.pkg, opt.ver, opt.date, opt.file); let tr = conn.transaction().unwrap(); tr.set_rollback(); |