This is an automated email from the ASF dual-hosted git repository.
github-bot pushed a commit to branch gh-pages
in repository https://gitbox.apache.org/repos/asf/iceberg-rust.git
The following commit(s) were added to refs/heads/gh-pages by this push:
new bba9e505 deploy: 835f528f4dd1d700158ce5339e14bc1da04626d1
bba9e505 is described below
commit bba9e505a6c1c3c84d2c45a041b46ac954831ace
Author: liurenjie1024 <[email protected]>
AuthorDate: Tue Nov 4 09:09:48 2025 +0000
deploy: 835f528f4dd1d700158ce5339e14bc1da04626d1
---
api/iceberg/arrow/delete_file_loader/index.html | 2 +-
.../arrow/caching_delete_file_loader.rs.html | 1006 +++++++++++---------
api/src/iceberg/arrow/delete_file_loader.rs.html | 138 +--
3 files changed, 626 insertions(+), 520 deletions(-)
diff --git a/api/iceberg/arrow/delete_file_loader/index.html
b/api/iceberg/arrow/delete_file_loader/index.html
index 7ac43aa6..0a68bc8a 100644
--- a/api/iceberg/arrow/delete_file_loader/index.html
+++ b/api/iceberg/arrow/delete_file_loader/index.html
@@ -1,2 +1,2 @@
-<!DOCTYPE html><html lang="en"><head><meta charset="utf-8"><meta
name="viewport" content="width=device-width, initial-scale=1.0"><meta
name="generator" content="rustdoc"><meta name="description" content="Delete
File loader"><title>iceberg::arrow::delete_file_loader -
Rust</title><script>if(window.location.protocol!=="file:")document.head.insertAdjacentHTML("beforeend","SourceSerif4-Regular-6b053e98.ttf.woff2,FiraSans-Italic-81dc35de.woff2,FiraSans-Regular-0fe48ade.woff2,FiraSans-MediumIt
[...]
+<!DOCTYPE html><html lang="en"><head><meta charset="utf-8"><meta
name="viewport" content="width=device-width, initial-scale=1.0"><meta
name="generator" content="rustdoc"><meta name="description" content="Delete
File loader"><title>iceberg::arrow::delete_file_loader -
Rust</title><script>if(window.location.protocol!=="file:")document.head.insertAdjacentHTML("beforeend","SourceSerif4-Regular-6b053e98.ttf.woff2,FiraSans-Italic-81dc35de.woff2,FiraSans-Regular-0fe48ade.woff2,FiraSans-MediumIt
[...]
</div></details><h2 id="traits" class="section-header">Traits<a href="#traits"
class="anchor">ยง</a></h2><dl class="item-table"><dt><a class="trait"
href="trait.DeleteFileLoader.html" title="trait
iceberg::arrow::delete_file_loader::DeleteFileLoader">Delete<wbr>File<wbr>Loader</a></dt><dd>Delete
File Loader</dd></dl></section></div></main></body></html>
\ No newline at end of file
diff --git a/api/src/iceberg/arrow/caching_delete_file_loader.rs.html
b/api/src/iceberg/arrow/caching_delete_file_loader.rs.html
index baa246c3..2fea5f78 100644
--- a/api/src/iceberg/arrow/caching_delete_file_loader.rs.html
+++ b/api/src/iceberg/arrow/caching_delete_file_loader.rs.html
@@ -224,466 +224,568 @@
<a href=#224 id=224 data-nosnippet>224</a> <span class="kw">let
</span>(sender, receiver) = channel();
<a href=#225 id=225 data-nosnippet>225</a>
del_filter.insert_equality_delete(<span
class="kw-2">&</span>task.file_path, receiver);
<a href=#226 id=226 data-nosnippet>226</a>
-<a href=#227 id=227 data-nosnippet>227</a> <span
class="prelude-val">Ok</span>(DeleteFileContext::FreshEqDel {
-<a href=#228 id=228 data-nosnippet>228</a> batch_stream:
BasicDeleteFileLoader::evolve_schema(
-<a href=#229 id=229 data-nosnippet>229</a>
basic_delete_file_loader
-<a href=#230 id=230 data-nosnippet>230</a>
.parquet_to_batch_stream(<span class="kw-2">&</span>task.file_path)
-<a href=#231 id=231 data-nosnippet>231</a> .<span
class="kw">await</span><span class="question-mark">?</span>,
-<a href=#232 id=232 data-nosnippet>232</a> schema,
-<a href=#233 id=233 data-nosnippet>233</a> )
-<a href=#234 id=234 data-nosnippet>234</a> .<span
class="kw">await</span><span class="question-mark">?</span>,
-<a href=#235 id=235 data-nosnippet>235</a> sender,
-<a href=#236 id=236 data-nosnippet>236</a> equality_ids:
HashSet::from_iter(task.equality_ids.clone().unwrap()),
-<a href=#237 id=237 data-nosnippet>237</a> })
-<a href=#238 id=238 data-nosnippet>238</a> }
-<a href=#239 id=239 data-nosnippet>239</a>
-<a href=#240 id=240 data-nosnippet>240</a> DataContentType::Data
=> <span class="prelude-val">Err</span>(Error::new(
-<a href=#241 id=241 data-nosnippet>241</a>
ErrorKind::Unexpected,
-<a href=#242 id=242 data-nosnippet>242</a> <span
class="string">"tasks with files of type Data not expected here"</span>,
-<a href=#243 id=243 data-nosnippet>243</a> )),
-<a href=#244 id=244 data-nosnippet>244</a> }
-<a href=#245 id=245 data-nosnippet>245</a> }
-<a href=#246 id=246 data-nosnippet>246</a>
-<a href=#247 id=247 data-nosnippet>247</a> <span class="kw">async fn
</span>parse_file_content_for_task(
-<a href=#248 id=248 data-nosnippet>248</a> ctx: DeleteFileContext,
-<a href=#249 id=249 data-nosnippet>249</a> ) -> <span
class="prelude-ty">Result</span><ParsedDeleteFileContext> {
-<a href=#250 id=250 data-nosnippet>250</a> <span class="kw">match
</span>ctx {
-<a href=#251 id=251 data-nosnippet>251</a>
DeleteFileContext::ExistingEqDel => <span
class="prelude-val">Ok</span>(ParsedDeleteFileContext::EqDel),
-<a href=#252 id=252 data-nosnippet>252</a>
DeleteFileContext::PosDels(batch_stream) => {
-<a href=#253 id=253 data-nosnippet>253</a> <span class="kw">let
</span>del_vecs =
-<a href=#254 id=254 data-nosnippet>254</a> <span
class="self">Self</span>::parse_positional_deletes_record_batch_stream(batch_stream).<span
class="kw">await</span><span class="question-mark">?</span>;
-<a href=#255 id=255 data-nosnippet>255</a> <span
class="prelude-val">Ok</span>(ParsedDeleteFileContext::DelVecs(del_vecs))
-<a href=#256 id=256 data-nosnippet>256</a> }
-<a href=#257 id=257 data-nosnippet>257</a>
DeleteFileContext::FreshEqDel {
-<a href=#258 id=258 data-nosnippet>258</a> sender,
-<a href=#259 id=259 data-nosnippet>259</a> batch_stream,
-<a href=#260 id=260 data-nosnippet>260</a> equality_ids,
-<a href=#261 id=261 data-nosnippet>261</a> } => {
-<a href=#262 id=262 data-nosnippet>262</a> <span class="kw">let
</span>predicate =
-<a href=#263 id=263 data-nosnippet>263</a> <span
class="self">Self</span>::parse_equality_deletes_record_batch_stream(batch_stream,
equality_ids)
-<a href=#264 id=264 data-nosnippet>264</a> .<span
class="kw">await</span><span class="question-mark">?</span>;
-<a href=#265 id=265 data-nosnippet>265</a>
-<a href=#266 id=266 data-nosnippet>266</a> sender
-<a href=#267 id=267 data-nosnippet>267</a> .send(predicate)
-<a href=#268 id=268 data-nosnippet>268</a> .map_err(|err| {
-<a href=#269 id=269 data-nosnippet>269</a> Error::new(
-<a href=#270 id=270 data-nosnippet>270</a>
ErrorKind::Unexpected,
-<a href=#271 id=271 data-nosnippet>271</a> <span
class="string">"Could not send eq delete predicate to state"</span>,
-<a href=#272 id=272 data-nosnippet>272</a> )
-<a href=#273 id=273 data-nosnippet>273</a> })
-<a href=#274 id=274 data-nosnippet>274</a> .map(|<span
class="kw">_</span>| ParsedDeleteFileContext::EqDel)
-<a href=#275 id=275 data-nosnippet>275</a> }
-<a href=#276 id=276 data-nosnippet>276</a> }
-<a href=#277 id=277 data-nosnippet>277</a> }
-<a href=#278 id=278 data-nosnippet>278</a>
-<a href=#279 id=279 data-nosnippet>279</a> <span class="doccomment">///
Parses a record batch stream coming from positional delete files
-<a href=#280 id=280 data-nosnippet>280</a> ///
-<a href=#281 id=281 data-nosnippet>281</a> /// Returns a map of data file
path to a delete vector
-<a href=#282 id=282 data-nosnippet>282</a> </span><span class="kw">async fn
</span>parse_positional_deletes_record_batch_stream(
-<a href=#283 id=283 data-nosnippet>283</a> <span class="kw-2">mut
</span>stream: ArrowRecordBatchStream,
-<a href=#284 id=284 data-nosnippet>284</a> ) -> <span
class="prelude-ty">Result</span><HashMap<String, DeleteVector>> {
-<a href=#285 id=285 data-nosnippet>285</a> <span class="kw">let
</span><span class="kw-2">mut </span>result: HashMap<String,
DeleteVector> = HashMap::default();
-<a href=#286 id=286 data-nosnippet>286</a>
-<a href=#287 id=287 data-nosnippet>287</a> <span class="kw">while let
</span><span class="prelude-val">Some</span>(batch) = stream.next().<span
class="kw">await </span>{
-<a href=#288 id=288 data-nosnippet>288</a> <span class="kw">let
</span>batch = batch<span class="question-mark">?</span>;
-<a href=#289 id=289 data-nosnippet>289</a> <span class="kw">let
</span>schema = batch.schema();
-<a href=#290 id=290 data-nosnippet>290</a> <span class="kw">let
</span>columns = batch.columns();
-<a href=#291 id=291 data-nosnippet>291</a>
-<a href=#292 id=292 data-nosnippet>292</a> <span class="kw">let
</span><span class="prelude-val">Some</span>(file_paths) = columns[<span
class="number">0</span>].as_any().downcast_ref::<StringArray>() <span
class="kw">else </span>{
-<a href=#293 id=293 data-nosnippet>293</a> <span
class="kw">return </span><span class="prelude-val">Err</span>(Error::new(
-<a href=#294 id=294 data-nosnippet>294</a>
ErrorKind::DataInvalid,
-<a href=#295 id=295 data-nosnippet>295</a> <span
class="string">"Could not downcast file paths array to StringArray"</span>,
-<a href=#296 id=296 data-nosnippet>296</a> ));
-<a href=#297 id=297 data-nosnippet>297</a> };
-<a href=#298 id=298 data-nosnippet>298</a> <span class="kw">let
</span><span class="prelude-val">Some</span>(positions) = columns[<span
class="number">1</span>].as_any().downcast_ref::<Int64Array>() <span
class="kw">else </span>{
+<a href=#227 id=227 data-nosnippet>227</a> <span
class="comment">// Per the Iceberg spec, evolve schema for equality deletes but
only for the
+<a href=#228 id=228 data-nosnippet>228</a> // equality_ids
columns, not all table columns.
+<a href=#229 id=229 data-nosnippet>229</a> </span><span
class="kw">let </span>equality_ids_vec = task.equality_ids.clone().unwrap();
+<a href=#230 id=230 data-nosnippet>230</a> <span class="kw">let
</span>evolved_stream = BasicDeleteFileLoader::evolve_schema(
+<a href=#231 id=231 data-nosnippet>231</a>
basic_delete_file_loader
+<a href=#232 id=232 data-nosnippet>232</a>
.parquet_to_batch_stream(<span class="kw-2">&</span>task.file_path)
+<a href=#233 id=233 data-nosnippet>233</a> .<span
class="kw">await</span><span class="question-mark">?</span>,
+<a href=#234 id=234 data-nosnippet>234</a> schema,
+<a href=#235 id=235 data-nosnippet>235</a> <span
class="kw-2">&</span>equality_ids_vec,
+<a href=#236 id=236 data-nosnippet>236</a> )
+<a href=#237 id=237 data-nosnippet>237</a> .<span
class="kw">await</span><span class="question-mark">?</span>;
+<a href=#238 id=238 data-nosnippet>238</a>
+<a href=#239 id=239 data-nosnippet>239</a> <span
class="prelude-val">Ok</span>(DeleteFileContext::FreshEqDel {
+<a href=#240 id=240 data-nosnippet>240</a> batch_stream:
evolved_stream,
+<a href=#241 id=241 data-nosnippet>241</a> sender,
+<a href=#242 id=242 data-nosnippet>242</a> equality_ids:
HashSet::from_iter(equality_ids_vec),
+<a href=#243 id=243 data-nosnippet>243</a> })
+<a href=#244 id=244 data-nosnippet>244</a> }
+<a href=#245 id=245 data-nosnippet>245</a>
+<a href=#246 id=246 data-nosnippet>246</a> DataContentType::Data
=> <span class="prelude-val">Err</span>(Error::new(
+<a href=#247 id=247 data-nosnippet>247</a>
ErrorKind::Unexpected,
+<a href=#248 id=248 data-nosnippet>248</a> <span
class="string">"tasks with files of type Data not expected here"</span>,
+<a href=#249 id=249 data-nosnippet>249</a> )),
+<a href=#250 id=250 data-nosnippet>250</a> }
+<a href=#251 id=251 data-nosnippet>251</a> }
+<a href=#252 id=252 data-nosnippet>252</a>
+<a href=#253 id=253 data-nosnippet>253</a> <span class="kw">async fn
</span>parse_file_content_for_task(
+<a href=#254 id=254 data-nosnippet>254</a> ctx: DeleteFileContext,
+<a href=#255 id=255 data-nosnippet>255</a> ) -> <span
class="prelude-ty">Result</span><ParsedDeleteFileContext> {
+<a href=#256 id=256 data-nosnippet>256</a> <span class="kw">match
</span>ctx {
+<a href=#257 id=257 data-nosnippet>257</a>
DeleteFileContext::ExistingEqDel => <span
class="prelude-val">Ok</span>(ParsedDeleteFileContext::EqDel),
+<a href=#258 id=258 data-nosnippet>258</a>
DeleteFileContext::PosDels(batch_stream) => {
+<a href=#259 id=259 data-nosnippet>259</a> <span class="kw">let
</span>del_vecs =
+<a href=#260 id=260 data-nosnippet>260</a> <span
class="self">Self</span>::parse_positional_deletes_record_batch_stream(batch_stream).<span
class="kw">await</span><span class="question-mark">?</span>;
+<a href=#261 id=261 data-nosnippet>261</a> <span
class="prelude-val">Ok</span>(ParsedDeleteFileContext::DelVecs(del_vecs))
+<a href=#262 id=262 data-nosnippet>262</a> }
+<a href=#263 id=263 data-nosnippet>263</a>
DeleteFileContext::FreshEqDel {
+<a href=#264 id=264 data-nosnippet>264</a> sender,
+<a href=#265 id=265 data-nosnippet>265</a> batch_stream,
+<a href=#266 id=266 data-nosnippet>266</a> equality_ids,
+<a href=#267 id=267 data-nosnippet>267</a> } => {
+<a href=#268 id=268 data-nosnippet>268</a> <span class="kw">let
</span>predicate =
+<a href=#269 id=269 data-nosnippet>269</a> <span
class="self">Self</span>::parse_equality_deletes_record_batch_stream(batch_stream,
equality_ids)
+<a href=#270 id=270 data-nosnippet>270</a> .<span
class="kw">await</span><span class="question-mark">?</span>;
+<a href=#271 id=271 data-nosnippet>271</a>
+<a href=#272 id=272 data-nosnippet>272</a> sender
+<a href=#273 id=273 data-nosnippet>273</a> .send(predicate)
+<a href=#274 id=274 data-nosnippet>274</a> .map_err(|err| {
+<a href=#275 id=275 data-nosnippet>275</a> Error::new(
+<a href=#276 id=276 data-nosnippet>276</a>
ErrorKind::Unexpected,
+<a href=#277 id=277 data-nosnippet>277</a> <span
class="string">"Could not send eq delete predicate to state"</span>,
+<a href=#278 id=278 data-nosnippet>278</a> )
+<a href=#279 id=279 data-nosnippet>279</a> })
+<a href=#280 id=280 data-nosnippet>280</a> .map(|<span
class="kw">_</span>| ParsedDeleteFileContext::EqDel)
+<a href=#281 id=281 data-nosnippet>281</a> }
+<a href=#282 id=282 data-nosnippet>282</a> }
+<a href=#283 id=283 data-nosnippet>283</a> }
+<a href=#284 id=284 data-nosnippet>284</a>
+<a href=#285 id=285 data-nosnippet>285</a> <span class="doccomment">///
Parses a record batch stream coming from positional delete files
+<a href=#286 id=286 data-nosnippet>286</a> ///
+<a href=#287 id=287 data-nosnippet>287</a> /// Returns a map of data file
path to a delete vector
+<a href=#288 id=288 data-nosnippet>288</a> </span><span class="kw">async fn
</span>parse_positional_deletes_record_batch_stream(
+<a href=#289 id=289 data-nosnippet>289</a> <span class="kw-2">mut
</span>stream: ArrowRecordBatchStream,
+<a href=#290 id=290 data-nosnippet>290</a> ) -> <span
class="prelude-ty">Result</span><HashMap<String, DeleteVector>> {
+<a href=#291 id=291 data-nosnippet>291</a> <span class="kw">let
</span><span class="kw-2">mut </span>result: HashMap<String,
DeleteVector> = HashMap::default();
+<a href=#292 id=292 data-nosnippet>292</a>
+<a href=#293 id=293 data-nosnippet>293</a> <span class="kw">while let
</span><span class="prelude-val">Some</span>(batch) = stream.next().<span
class="kw">await </span>{
+<a href=#294 id=294 data-nosnippet>294</a> <span class="kw">let
</span>batch = batch<span class="question-mark">?</span>;
+<a href=#295 id=295 data-nosnippet>295</a> <span class="kw">let
</span>schema = batch.schema();
+<a href=#296 id=296 data-nosnippet>296</a> <span class="kw">let
</span>columns = batch.columns();
+<a href=#297 id=297 data-nosnippet>297</a>
+<a href=#298 id=298 data-nosnippet>298</a> <span class="kw">let
</span><span class="prelude-val">Some</span>(file_paths) = columns[<span
class="number">0</span>].as_any().downcast_ref::<StringArray>() <span
class="kw">else </span>{
<a href=#299 id=299 data-nosnippet>299</a> <span
class="kw">return </span><span class="prelude-val">Err</span>(Error::new(
<a href=#300 id=300 data-nosnippet>300</a>
ErrorKind::DataInvalid,
-<a href=#301 id=301 data-nosnippet>301</a> <span
class="string">"Could not downcast positions array to Int64Array"</span>,
+<a href=#301 id=301 data-nosnippet>301</a> <span
class="string">"Could not downcast file paths array to StringArray"</span>,
<a href=#302 id=302 data-nosnippet>302</a> ));
<a href=#303 id=303 data-nosnippet>303</a> };
-<a href=#304 id=304 data-nosnippet>304</a>
-<a href=#305 id=305 data-nosnippet>305</a> <span class="kw">for
</span>(file_path, pos) <span class="kw">in
</span>file_paths.iter().zip(positions.iter()) {
-<a href=#306 id=306 data-nosnippet>306</a> <span class="kw">let
</span>(<span class="prelude-val">Some</span>(file_path), <span
class="prelude-val">Some</span>(pos)) = (file_path, pos) <span class="kw">else
</span>{
-<a href=#307 id=307 data-nosnippet>307</a> <span
class="kw">return </span><span class="prelude-val">Err</span>(Error::new(
-<a href=#308 id=308 data-nosnippet>308</a>
ErrorKind::DataInvalid,
-<a href=#309 id=309 data-nosnippet>309</a> <span
class="string">"null values in delete file"</span>,
-<a href=#310 id=310 data-nosnippet>310</a> ));
-<a href=#311 id=311 data-nosnippet>311</a> };
-<a href=#312 id=312 data-nosnippet>312</a>
-<a href=#313 id=313 data-nosnippet>313</a> result
-<a href=#314 id=314 data-nosnippet>314</a>
.entry(file_path.to_string())
-<a href=#315 id=315 data-nosnippet>315</a> .or_default()
-<a href=#316 id=316 data-nosnippet>316</a> .insert(pos
<span class="kw">as </span>u64);
-<a href=#317 id=317 data-nosnippet>317</a> }
-<a href=#318 id=318 data-nosnippet>318</a> }
-<a href=#319 id=319 data-nosnippet>319</a>
-<a href=#320 id=320 data-nosnippet>320</a> <span
class="prelude-val">Ok</span>(result)
-<a href=#321 id=321 data-nosnippet>321</a> }
-<a href=#322 id=322 data-nosnippet>322</a>
-<a href=#323 id=323 data-nosnippet>323</a> <span class="kw">async fn
</span>parse_equality_deletes_record_batch_stream(
-<a href=#324 id=324 data-nosnippet>324</a> <span class="kw-2">mut
</span>stream: ArrowRecordBatchStream,
-<a href=#325 id=325 data-nosnippet>325</a> equality_ids:
HashSet<i32>,
-<a href=#326 id=326 data-nosnippet>326</a> ) -> <span
class="prelude-ty">Result</span><Predicate> {
-<a href=#327 id=327 data-nosnippet>327</a> <span class="kw">let
</span><span class="kw-2">mut </span>result_predicate = AlwaysTrue;
-<a href=#328 id=328 data-nosnippet>328</a> <span class="kw">let
</span><span class="kw-2">mut </span>batch_schema_iceberg: <span
class="prelude-ty">Option</span><Schema> = <span
class="prelude-val">None</span>;
-<a href=#329 id=329 data-nosnippet>329</a> <span class="kw">let
</span>accessor = EqDelRecordBatchPartnerAccessor;
-<a href=#330 id=330 data-nosnippet>330</a>
-<a href=#331 id=331 data-nosnippet>331</a> <span class="kw">while let
</span><span class="prelude-val">Some</span>(record_batch) =
stream.next().<span class="kw">await </span>{
-<a href=#332 id=332 data-nosnippet>332</a> <span class="kw">let
</span>record_batch = record_batch<span class="question-mark">?</span>;
-<a href=#333 id=333 data-nosnippet>333</a>
-<a href=#334 id=334 data-nosnippet>334</a> <span class="kw">if
</span>record_batch.num_columns() == <span class="number">0 </span>{
-<a href=#335 id=335 data-nosnippet>335</a> <span
class="kw">return </span><span class="prelude-val">Ok</span>(AlwaysTrue);
-<a href=#336 id=336 data-nosnippet>336</a> }
-<a href=#337 id=337 data-nosnippet>337</a>
-<a href=#338 id=338 data-nosnippet>338</a> <span class="kw">let
</span>schema = <span class="kw">match </span><span
class="kw-2">&</span>batch_schema_iceberg {
-<a href=#339 id=339 data-nosnippet>339</a> <span
class="prelude-val">Some</span>(schema) => schema,
-<a href=#340 id=340 data-nosnippet>340</a> <span
class="prelude-val">None </span>=> {
-<a href=#341 id=341 data-nosnippet>341</a> <span
class="kw">let </span>schema =
arrow_schema_to_schema(record_batch.schema().as_ref())<span
class="question-mark">?</span>;
-<a href=#342 id=342 data-nosnippet>342</a>
batch_schema_iceberg = <span class="prelude-val">Some</span>(schema);
-<a href=#343 id=343 data-nosnippet>343</a>
batch_schema_iceberg.as_ref().unwrap()
-<a href=#344 id=344 data-nosnippet>344</a> }
-<a href=#345 id=345 data-nosnippet>345</a> };
-<a href=#346 id=346 data-nosnippet>346</a>
-<a href=#347 id=347 data-nosnippet>347</a> <span class="kw">let
</span>root_array: ArrayRef = Arc::new(StructArray::from(record_batch));
-<a href=#348 id=348 data-nosnippet>348</a>
-<a href=#349 id=349 data-nosnippet>349</a> <span class="kw">let
</span><span class="kw-2">mut </span>processor =
EqDelColumnProcessor::new(<span class="kw-2">&</span>equality_ids);
-<a href=#350 id=350 data-nosnippet>350</a>
visit_schema_with_partner(schema, <span class="kw-2">&</span>root_array,
<span class="kw-2">&mut </span>processor, <span
class="kw-2">&</span>accessor)<span class="question-mark">?</span>;
-<a href=#351 id=351 data-nosnippet>351</a>
-<a href=#352 id=352 data-nosnippet>352</a> <span class="kw">let
</span><span class="kw-2">mut </span>datum_columns_with_names =
processor.finish()<span class="question-mark">?</span>;
-<a href=#353 id=353 data-nosnippet>353</a> <span class="kw">if
</span>datum_columns_with_names.is_empty() {
-<a href=#354 id=354 data-nosnippet>354</a> <span
class="kw">continue</span>;
-<a href=#355 id=355 data-nosnippet>355</a> }
-<a href=#356 id=356 data-nosnippet>356</a>
-<a href=#357 id=357 data-nosnippet>357</a> <span class="comment">//
Process the collected columns in lockstep
-<a href=#358 id=358 data-nosnippet>358</a> </span><span
class="attr">#[allow(clippy::len_zero)]
-<a href=#359 id=359 data-nosnippet>359</a> </span><span
class="kw">while </span>datum_columns_with_names[<span
class="number">0</span>].<span class="number">0</span>.len() > <span
class="number">0 </span>{
-<a href=#360 id=360 data-nosnippet>360</a> <span class="kw">let
</span><span class="kw-2">mut </span>row_predicate = AlwaysTrue;
-<a href=#361 id=361 data-nosnippet>361</a> <span class="kw">for
</span><span class="kw-2">&mut </span>(<span class="kw-2">ref mut
</span>column, <span class="kw-2">ref </span>field_name) <span class="kw">in
</span><span class="kw-2">&mut </span>datum_columns_with_names {
-<a href=#362 id=362 data-nosnippet>362</a> <span
class="kw">if let </span><span class="prelude-val">Some</span>(item) =
column.next() {
-<a href=#363 id=363 data-nosnippet>363</a> <span
class="kw">let </span>cell_predicate = <span class="kw">if let </span><span
class="prelude-val">Some</span>(datum) = item<span class="question-mark">?
</span>{
-<a href=#364 id=364 data-nosnippet>364</a>
Reference::new(field_name.clone()).equal_to(datum.clone())
-<a href=#365 id=365 data-nosnippet>365</a> } <span
class="kw">else </span>{
-<a href=#366 id=366 data-nosnippet>366</a>
Reference::new(field_name.clone()).is_null()
-<a href=#367 id=367 data-nosnippet>367</a> };
-<a href=#368 id=368 data-nosnippet>368</a>
row_predicate = row_predicate.and(cell_predicate)
-<a href=#369 id=369 data-nosnippet>369</a> }
-<a href=#370 id=370 data-nosnippet>370</a> }
-<a href=#371 id=371 data-nosnippet>371</a> result_predicate =
result_predicate.and(row_predicate.not());
-<a href=#372 id=372 data-nosnippet>372</a> }
-<a href=#373 id=373 data-nosnippet>373</a> }
-<a href=#374 id=374 data-nosnippet>374</a> <span
class="prelude-val">Ok</span>(result_predicate.rewrite_not())
-<a href=#375 id=375 data-nosnippet>375</a> }
-<a href=#376 id=376 data-nosnippet>376</a>}
-<a href=#377 id=377 data-nosnippet>377</a>
-<a href=#378 id=378 data-nosnippet>378</a><span class="kw">struct
</span>EqDelColumnProcessor<<span class="lifetime">'a</span>> {
-<a href=#379 id=379 data-nosnippet>379</a> equality_ids: <span
class="kw-2">&</span><span class="lifetime">'a </span>HashSet<i32>,
-<a href=#380 id=380 data-nosnippet>380</a> collected_columns:
Vec<(ArrayRef, String, Type)>,
-<a href=#381 id=381 data-nosnippet>381</a>}
-<a href=#382 id=382 data-nosnippet>382</a>
-<a href=#383 id=383 data-nosnippet>383</a><span
class="kw">impl</span><<span class="lifetime">'a</span>>
EqDelColumnProcessor<<span class="lifetime">'a</span>> {
-<a href=#384 id=384 data-nosnippet>384</a> <span class="kw">fn
</span>new(equality_ids: <span class="kw-2">&</span><span
class="lifetime">'a </span>HashSet<i32>) -> <span class="self">Self
</span>{
-<a href=#385 id=385 data-nosnippet>385</a> <span class="self">Self
</span>{
-<a href=#386 id=386 data-nosnippet>386</a> equality_ids,
-<a href=#387 id=387 data-nosnippet>387</a> collected_columns:
Vec::with_capacity(equality_ids.len()),
-<a href=#388 id=388 data-nosnippet>388</a> }
-<a href=#389 id=389 data-nosnippet>389</a> }
-<a href=#390 id=390 data-nosnippet>390</a>
-<a href=#391 id=391 data-nosnippet>391</a> <span
class="attr">#[allow(clippy::type_complexity)]
-<a href=#392 id=392 data-nosnippet>392</a> </span><span class="kw">fn
</span>finish(
-<a href=#393 id=393 data-nosnippet>393</a> <span
class="self">self</span>,
-<a href=#394 id=394 data-nosnippet>394</a> ) -> <span
class="prelude-ty">Result</span><
-<a href=#395 id=395 data-nosnippet>395</a> Vec<(
-<a href=#396 id=396 data-nosnippet>396</a> Box<<span
class="kw">dyn </span>ExactSizeIterator<Item = <span
class="prelude-ty">Result</span><<span
class="prelude-ty">Option</span><Datum>>>>,
-<a href=#397 id=397 data-nosnippet>397</a> String,
-<a href=#398 id=398 data-nosnippet>398</a> )>,
-<a href=#399 id=399 data-nosnippet>399</a> > {
-<a href=#400 id=400 data-nosnippet>400</a> <span
class="self">self</span>.collected_columns
-<a href=#401 id=401 data-nosnippet>401</a> .into_iter()
-<a href=#402 id=402 data-nosnippet>402</a> .map(|(array,
field_name, field_type)| {
-<a href=#403 id=403 data-nosnippet>403</a> <span class="kw">let
</span>primitive_type = field_type
-<a href=#404 id=404 data-nosnippet>404</a>
.as_primitive_type()
-<a href=#405 id=405 data-nosnippet>405</a> .ok_or_else(|| {
-<a href=#406 id=406 data-nosnippet>406</a>
Error::new(ErrorKind::Unexpected, <span class="string">"field is not a
primitive type"</span>)
-<a href=#407 id=407 data-nosnippet>407</a> })<span
class="question-mark">?
-<a href=#408 id=408 data-nosnippet>408</a> </span>.clone();
-<a href=#409 id=409 data-nosnippet>409</a>
-<a href=#410 id=410 data-nosnippet>410</a> <span class="kw">let
</span>lit_vec = arrow_primitive_to_literal(<span
class="kw-2">&</span>array, <span class="kw-2">&</span>field_type)<span
class="question-mark">?</span>;
-<a href=#411 id=411 data-nosnippet>411</a> <span class="kw">let
</span>datum_iterator: Box<<span class="kw">dyn
</span>ExactSizeIterator<Item = <span
class="prelude-ty">Result</span><<span
class="prelude-ty">Option</span><Datum>>>> =
-<a href=#412 id=412 data-nosnippet>412</a>
Box::new(lit_vec.into_iter().map(<span class="kw">move </span>|c| {
-<a href=#413 id=413 data-nosnippet>413</a>
c.map(|literal| {
-<a href=#414 id=414 data-nosnippet>414</a> literal
-<a href=#415 id=415 data-nosnippet>415</a>
.as_primitive_literal()
-<a href=#416 id=416 data-nosnippet>416</a>
.map(|primitive_literal| {
-<a href=#417 id=417 data-nosnippet>417</a>
Datum::new(primitive_type.clone(), primitive_literal)
-<a href=#418 id=418 data-nosnippet>418</a> })
-<a href=#419 id=419 data-nosnippet>419</a>
.ok_or(Error::new(
-<a href=#420 id=420 data-nosnippet>420</a>
ErrorKind::Unexpected,
-<a href=#421 id=421 data-nosnippet>421</a>
<span class="string">"failed to convert to primitive literal"</span>,
-<a href=#422 id=422 data-nosnippet>422</a> ))
-<a href=#423 id=423 data-nosnippet>423</a> })
-<a href=#424 id=424 data-nosnippet>424</a> .transpose()
-<a href=#425 id=425 data-nosnippet>425</a> }));
-<a href=#426 id=426 data-nosnippet>426</a>
-<a href=#427 id=427 data-nosnippet>427</a> <span
class="prelude-val">Ok</span>((datum_iterator, field_name))
-<a href=#428 id=428 data-nosnippet>428</a> })
-<a href=#429 id=429 data-nosnippet>429</a> .collect::<<span
class="prelude-ty">Result</span><Vec<<span
class="kw">_</span>>>>()
-<a href=#430 id=430 data-nosnippet>430</a> }
-<a href=#431 id=431 data-nosnippet>431</a>}
+<a href=#304 id=304 data-nosnippet>304</a> <span class="kw">let
</span><span class="prelude-val">Some</span>(positions) = columns[<span
class="number">1</span>].as_any().downcast_ref::<Int64Array>() <span
class="kw">else </span>{
+<a href=#305 id=305 data-nosnippet>305</a> <span
class="kw">return </span><span class="prelude-val">Err</span>(Error::new(
+<a href=#306 id=306 data-nosnippet>306</a>
ErrorKind::DataInvalid,
+<a href=#307 id=307 data-nosnippet>307</a> <span
class="string">"Could not downcast positions array to Int64Array"</span>,
+<a href=#308 id=308 data-nosnippet>308</a> ));
+<a href=#309 id=309 data-nosnippet>309</a> };
+<a href=#310 id=310 data-nosnippet>310</a>
+<a href=#311 id=311 data-nosnippet>311</a> <span class="kw">for
</span>(file_path, pos) <span class="kw">in
</span>file_paths.iter().zip(positions.iter()) {
+<a href=#312 id=312 data-nosnippet>312</a> <span class="kw">let
</span>(<span class="prelude-val">Some</span>(file_path), <span
class="prelude-val">Some</span>(pos)) = (file_path, pos) <span class="kw">else
</span>{
+<a href=#313 id=313 data-nosnippet>313</a> <span
class="kw">return </span><span class="prelude-val">Err</span>(Error::new(
+<a href=#314 id=314 data-nosnippet>314</a>
ErrorKind::DataInvalid,
+<a href=#315 id=315 data-nosnippet>315</a> <span
class="string">"null values in delete file"</span>,
+<a href=#316 id=316 data-nosnippet>316</a> ));
+<a href=#317 id=317 data-nosnippet>317</a> };
+<a href=#318 id=318 data-nosnippet>318</a>
+<a href=#319 id=319 data-nosnippet>319</a> result
+<a href=#320 id=320 data-nosnippet>320</a>
.entry(file_path.to_string())
+<a href=#321 id=321 data-nosnippet>321</a> .or_default()
+<a href=#322 id=322 data-nosnippet>322</a> .insert(pos
<span class="kw">as </span>u64);
+<a href=#323 id=323 data-nosnippet>323</a> }
+<a href=#324 id=324 data-nosnippet>324</a> }
+<a href=#325 id=325 data-nosnippet>325</a>
+<a href=#326 id=326 data-nosnippet>326</a> <span
class="prelude-val">Ok</span>(result)
+<a href=#327 id=327 data-nosnippet>327</a> }
+<a href=#328 id=328 data-nosnippet>328</a>
+<a href=#329 id=329 data-nosnippet>329</a> <span class="kw">async fn
</span>parse_equality_deletes_record_batch_stream(
+<a href=#330 id=330 data-nosnippet>330</a> <span class="kw-2">mut
</span>stream: ArrowRecordBatchStream,
+<a href=#331 id=331 data-nosnippet>331</a> equality_ids:
HashSet<i32>,
+<a href=#332 id=332 data-nosnippet>332</a> ) -> <span
class="prelude-ty">Result</span><Predicate> {
+<a href=#333 id=333 data-nosnippet>333</a> <span class="kw">let
</span><span class="kw-2">mut </span>result_predicate = AlwaysTrue;
+<a href=#334 id=334 data-nosnippet>334</a> <span class="kw">let
</span><span class="kw-2">mut </span>batch_schema_iceberg: <span
class="prelude-ty">Option</span><Schema> = <span
class="prelude-val">None</span>;
+<a href=#335 id=335 data-nosnippet>335</a> <span class="kw">let
</span>accessor = EqDelRecordBatchPartnerAccessor;
+<a href=#336 id=336 data-nosnippet>336</a>
+<a href=#337 id=337 data-nosnippet>337</a> <span class="kw">while let
</span><span class="prelude-val">Some</span>(record_batch) =
stream.next().<span class="kw">await </span>{
+<a href=#338 id=338 data-nosnippet>338</a> <span class="kw">let
</span>record_batch = record_batch<span class="question-mark">?</span>;
+<a href=#339 id=339 data-nosnippet>339</a>
+<a href=#340 id=340 data-nosnippet>340</a> <span class="kw">if
</span>record_batch.num_columns() == <span class="number">0 </span>{
+<a href=#341 id=341 data-nosnippet>341</a> <span
class="kw">return </span><span class="prelude-val">Ok</span>(AlwaysTrue);
+<a href=#342 id=342 data-nosnippet>342</a> }
+<a href=#343 id=343 data-nosnippet>343</a>
+<a href=#344 id=344 data-nosnippet>344</a> <span class="kw">let
</span>schema = <span class="kw">match </span><span
class="kw-2">&</span>batch_schema_iceberg {
+<a href=#345 id=345 data-nosnippet>345</a> <span
class="prelude-val">Some</span>(schema) => schema,
+<a href=#346 id=346 data-nosnippet>346</a> <span
class="prelude-val">None </span>=> {
+<a href=#347 id=347 data-nosnippet>347</a> <span
class="kw">let </span>schema =
arrow_schema_to_schema(record_batch.schema().as_ref())<span
class="question-mark">?</span>;
+<a href=#348 id=348 data-nosnippet>348</a>
batch_schema_iceberg = <span class="prelude-val">Some</span>(schema);
+<a href=#349 id=349 data-nosnippet>349</a>
batch_schema_iceberg.as_ref().unwrap()
+<a href=#350 id=350 data-nosnippet>350</a> }
+<a href=#351 id=351 data-nosnippet>351</a> };
+<a href=#352 id=352 data-nosnippet>352</a>
+<a href=#353 id=353 data-nosnippet>353</a> <span class="kw">let
</span>root_array: ArrayRef = Arc::new(StructArray::from(record_batch));
+<a href=#354 id=354 data-nosnippet>354</a>
+<a href=#355 id=355 data-nosnippet>355</a> <span class="kw">let
</span><span class="kw-2">mut </span>processor =
EqDelColumnProcessor::new(<span class="kw-2">&</span>equality_ids);
+<a href=#356 id=356 data-nosnippet>356</a>
visit_schema_with_partner(schema, <span class="kw-2">&</span>root_array,
<span class="kw-2">&mut </span>processor, <span
class="kw-2">&</span>accessor)<span class="question-mark">?</span>;
+<a href=#357 id=357 data-nosnippet>357</a>
+<a href=#358 id=358 data-nosnippet>358</a> <span class="kw">let
</span><span class="kw-2">mut </span>datum_columns_with_names =
processor.finish()<span class="question-mark">?</span>;
+<a href=#359 id=359 data-nosnippet>359</a> <span class="kw">if
</span>datum_columns_with_names.is_empty() {
+<a href=#360 id=360 data-nosnippet>360</a> <span
class="kw">continue</span>;
+<a href=#361 id=361 data-nosnippet>361</a> }
+<a href=#362 id=362 data-nosnippet>362</a>
+<a href=#363 id=363 data-nosnippet>363</a> <span class="comment">//
Process the collected columns in lockstep
+<a href=#364 id=364 data-nosnippet>364</a> </span><span
class="attr">#[allow(clippy::len_zero)]
+<a href=#365 id=365 data-nosnippet>365</a> </span><span
class="kw">while </span>datum_columns_with_names[<span
class="number">0</span>].<span class="number">0</span>.len() > <span
class="number">0 </span>{
+<a href=#366 id=366 data-nosnippet>366</a> <span class="kw">let
</span><span class="kw-2">mut </span>row_predicate = AlwaysTrue;
+<a href=#367 id=367 data-nosnippet>367</a> <span class="kw">for
</span><span class="kw-2">&mut </span>(<span class="kw-2">ref mut
</span>column, <span class="kw-2">ref </span>field_name) <span class="kw">in
</span><span class="kw-2">&mut </span>datum_columns_with_names {
+<a href=#368 id=368 data-nosnippet>368</a> <span
class="kw">if let </span><span class="prelude-val">Some</span>(item) =
column.next() {
+<a href=#369 id=369 data-nosnippet>369</a> <span
class="kw">let </span>cell_predicate = <span class="kw">if let </span><span
class="prelude-val">Some</span>(datum) = item<span class="question-mark">?
</span>{
+<a href=#370 id=370 data-nosnippet>370</a>
Reference::new(field_name.clone()).equal_to(datum.clone())
+<a href=#371 id=371 data-nosnippet>371</a> } <span
class="kw">else </span>{
+<a href=#372 id=372 data-nosnippet>372</a>
Reference::new(field_name.clone()).is_null()
+<a href=#373 id=373 data-nosnippet>373</a> };
+<a href=#374 id=374 data-nosnippet>374</a>
row_predicate = row_predicate.and(cell_predicate)
+<a href=#375 id=375 data-nosnippet>375</a> }
+<a href=#376 id=376 data-nosnippet>376</a> }
+<a href=#377 id=377 data-nosnippet>377</a> result_predicate =
result_predicate.and(row_predicate.not());
+<a href=#378 id=378 data-nosnippet>378</a> }
+<a href=#379 id=379 data-nosnippet>379</a> }
+<a href=#380 id=380 data-nosnippet>380</a> <span
class="prelude-val">Ok</span>(result_predicate.rewrite_not())
+<a href=#381 id=381 data-nosnippet>381</a> }
+<a href=#382 id=382 data-nosnippet>382</a>}
+<a href=#383 id=383 data-nosnippet>383</a>
+<a href=#384 id=384 data-nosnippet>384</a><span class="kw">struct
</span>EqDelColumnProcessor<<span class="lifetime">'a</span>> {
+<a href=#385 id=385 data-nosnippet>385</a> equality_ids: <span
class="kw-2">&</span><span class="lifetime">'a </span>HashSet<i32>,
+<a href=#386 id=386 data-nosnippet>386</a> collected_columns:
Vec<(ArrayRef, String, Type)>,
+<a href=#387 id=387 data-nosnippet>387</a>}
+<a href=#388 id=388 data-nosnippet>388</a>
+<a href=#389 id=389 data-nosnippet>389</a><span
class="kw">impl</span><<span class="lifetime">'a</span>>
EqDelColumnProcessor<<span class="lifetime">'a</span>> {
+<a href=#390 id=390 data-nosnippet>390</a> <span class="kw">fn
</span>new(equality_ids: <span class="kw-2">&</span><span
class="lifetime">'a </span>HashSet<i32>) -> <span class="self">Self
</span>{
+<a href=#391 id=391 data-nosnippet>391</a> <span class="self">Self
</span>{
+<a href=#392 id=392 data-nosnippet>392</a> equality_ids,
+<a href=#393 id=393 data-nosnippet>393</a> collected_columns:
Vec::with_capacity(equality_ids.len()),
+<a href=#394 id=394 data-nosnippet>394</a> }
+<a href=#395 id=395 data-nosnippet>395</a> }
+<a href=#396 id=396 data-nosnippet>396</a>
+<a href=#397 id=397 data-nosnippet>397</a> <span
class="attr">#[allow(clippy::type_complexity)]
+<a href=#398 id=398 data-nosnippet>398</a> </span><span class="kw">fn
</span>finish(
+<a href=#399 id=399 data-nosnippet>399</a> <span
class="self">self</span>,
+<a href=#400 id=400 data-nosnippet>400</a> ) -> <span
class="prelude-ty">Result</span><
+<a href=#401 id=401 data-nosnippet>401</a> Vec<(
+<a href=#402 id=402 data-nosnippet>402</a> Box<<span
class="kw">dyn </span>ExactSizeIterator<Item = <span
class="prelude-ty">Result</span><<span
class="prelude-ty">Option</span><Datum>>>>,
+<a href=#403 id=403 data-nosnippet>403</a> String,
+<a href=#404 id=404 data-nosnippet>404</a> )>,
+<a href=#405 id=405 data-nosnippet>405</a> > {
+<a href=#406 id=406 data-nosnippet>406</a> <span
class="self">self</span>.collected_columns
+<a href=#407 id=407 data-nosnippet>407</a> .into_iter()
+<a href=#408 id=408 data-nosnippet>408</a> .map(|(array,
field_name, field_type)| {
+<a href=#409 id=409 data-nosnippet>409</a> <span class="kw">let
</span>primitive_type = field_type
+<a href=#410 id=410 data-nosnippet>410</a>
.as_primitive_type()
+<a href=#411 id=411 data-nosnippet>411</a> .ok_or_else(|| {
+<a href=#412 id=412 data-nosnippet>412</a>
Error::new(ErrorKind::Unexpected, <span class="string">"field is not a
primitive type"</span>)
+<a href=#413 id=413 data-nosnippet>413</a> })<span
class="question-mark">?
+<a href=#414 id=414 data-nosnippet>414</a> </span>.clone();
+<a href=#415 id=415 data-nosnippet>415</a>
+<a href=#416 id=416 data-nosnippet>416</a> <span class="kw">let
</span>lit_vec = arrow_primitive_to_literal(<span
class="kw-2">&</span>array, <span class="kw-2">&</span>field_type)<span
class="question-mark">?</span>;
+<a href=#417 id=417 data-nosnippet>417</a> <span class="kw">let
</span>datum_iterator: Box<<span class="kw">dyn
</span>ExactSizeIterator<Item = <span
class="prelude-ty">Result</span><<span
class="prelude-ty">Option</span><Datum>>>> =
+<a href=#418 id=418 data-nosnippet>418</a>
Box::new(lit_vec.into_iter().map(<span class="kw">move </span>|c| {
+<a href=#419 id=419 data-nosnippet>419</a>
c.map(|literal| {
+<a href=#420 id=420 data-nosnippet>420</a> literal
+<a href=#421 id=421 data-nosnippet>421</a>
.as_primitive_literal()
+<a href=#422 id=422 data-nosnippet>422</a>
.map(|primitive_literal| {
+<a href=#423 id=423 data-nosnippet>423</a>
Datum::new(primitive_type.clone(), primitive_literal)
+<a href=#424 id=424 data-nosnippet>424</a> })
+<a href=#425 id=425 data-nosnippet>425</a>
.ok_or(Error::new(
+<a href=#426 id=426 data-nosnippet>426</a>
ErrorKind::Unexpected,
+<a href=#427 id=427 data-nosnippet>427</a>
<span class="string">"failed to convert to primitive literal"</span>,
+<a href=#428 id=428 data-nosnippet>428</a> ))
+<a href=#429 id=429 data-nosnippet>429</a> })
+<a href=#430 id=430 data-nosnippet>430</a> .transpose()
+<a href=#431 id=431 data-nosnippet>431</a> }));
<a href=#432 id=432 data-nosnippet>432</a>
-<a href=#433 id=433 data-nosnippet>433</a><span class="kw">impl
</span>SchemaWithPartnerVisitor<ArrayRef> <span class="kw">for
</span>EqDelColumnProcessor<<span class="lifetime">'_</span>> {
-<a href=#434 id=434 data-nosnippet>434</a> <span class="kw">type </span>T =
();
-<a href=#435 id=435 data-nosnippet>435</a>
-<a href=#436 id=436 data-nosnippet>436</a> <span class="kw">fn
</span>schema(<span class="kw-2">&mut </span><span
class="self">self</span>, _schema: <span class="kw-2">&</span>Schema,
_partner: <span class="kw-2">&</span>ArrayRef, _value: ()) -> <span
class="prelude-ty">Result</span><()> {
-<a href=#437 id=437 data-nosnippet>437</a> <span
class="prelude-val">Ok</span>(())
-<a href=#438 id=438 data-nosnippet>438</a> }
-<a href=#439 id=439 data-nosnippet>439</a>
-<a href=#440 id=440 data-nosnippet>440</a> <span class="kw">fn
</span>field(<span class="kw-2">&mut </span><span class="self">self</span>,
field: <span class="kw-2">&</span>NestedFieldRef, partner: <span
class="kw-2">&</span>ArrayRef, _value: ()) -> <span
class="prelude-ty">Result</span><()> {
-<a href=#441 id=441 data-nosnippet>441</a> <span class="kw">if
</span><span class="self">self</span>.equality_ids.contains(<span
class="kw-2">&</span>field.id) &&
field.field_type.as_primitive_type().is_some() {
-<a href=#442 id=442 data-nosnippet>442</a> <span
class="self">self</span>.collected_columns.push((
-<a href=#443 id=443 data-nosnippet>443</a> partner.clone(),
-<a href=#444 id=444 data-nosnippet>444</a> field.name.clone(),
-<a href=#445 id=445 data-nosnippet>445</a>
field.field_type.as_ref().clone(),
-<a href=#446 id=446 data-nosnippet>446</a> ));
-<a href=#447 id=447 data-nosnippet>447</a> }
-<a href=#448 id=448 data-nosnippet>448</a> <span
class="prelude-val">Ok</span>(())
-<a href=#449 id=449 data-nosnippet>449</a> }
-<a href=#450 id=450 data-nosnippet>450</a>
-<a href=#451 id=451 data-nosnippet>451</a> <span class="kw">fn
</span>r#struct(
-<a href=#452 id=452 data-nosnippet>452</a> <span class="kw-2">&mut
</span><span class="self">self</span>,
-<a href=#453 id=453 data-nosnippet>453</a> _struct: <span
class="kw-2">&</span>StructType,
-<a href=#454 id=454 data-nosnippet>454</a> _partner: <span
class="kw-2">&</span>ArrayRef,
-<a href=#455 id=455 data-nosnippet>455</a> _results: Vec<()>,
-<a href=#456 id=456 data-nosnippet>456</a> ) -> <span
class="prelude-ty">Result</span><()> {
-<a href=#457 id=457 data-nosnippet>457</a> <span
class="prelude-val">Ok</span>(())
-<a href=#458 id=458 data-nosnippet>458</a> }
-<a href=#459 id=459 data-nosnippet>459</a>
-<a href=#460 id=460 data-nosnippet>460</a> <span class="kw">fn
</span>list(<span class="kw-2">&mut </span><span class="self">self</span>,
_list: <span class="kw-2">&</span>ListType, _partner: <span
class="kw-2">&</span>ArrayRef, _value: ()) -> <span
class="prelude-ty">Result</span><()> {
-<a href=#461 id=461 data-nosnippet>461</a> <span
class="prelude-val">Ok</span>(())
-<a href=#462 id=462 data-nosnippet>462</a> }
-<a href=#463 id=463 data-nosnippet>463</a>
-<a href=#464 id=464 data-nosnippet>464</a> <span class="kw">fn </span>map(
-<a href=#465 id=465 data-nosnippet>465</a> <span class="kw-2">&mut
</span><span class="self">self</span>,
-<a href=#466 id=466 data-nosnippet>466</a> _map: <span
class="kw-2">&</span>MapType,
-<a href=#467 id=467 data-nosnippet>467</a> _partner: <span
class="kw-2">&</span>ArrayRef,
-<a href=#468 id=468 data-nosnippet>468</a> _key_value: (),
-<a href=#469 id=469 data-nosnippet>469</a> _value: (),
-<a href=#470 id=470 data-nosnippet>470</a> ) -> <span
class="prelude-ty">Result</span><()> {
-<a href=#471 id=471 data-nosnippet>471</a> <span
class="prelude-val">Ok</span>(())
-<a href=#472 id=472 data-nosnippet>472</a> }
-<a href=#473 id=473 data-nosnippet>473</a>
-<a href=#474 id=474 data-nosnippet>474</a> <span class="kw">fn
</span>primitive(<span class="kw-2">&mut </span><span
class="self">self</span>, _primitive: <span
class="kw-2">&</span>PrimitiveType, _partner: <span
class="kw-2">&</span>ArrayRef) -> <span
class="prelude-ty">Result</span><()> {
-<a href=#475 id=475 data-nosnippet>475</a> <span
class="prelude-val">Ok</span>(())
-<a href=#476 id=476 data-nosnippet>476</a> }
-<a href=#477 id=477 data-nosnippet>477</a>}
-<a href=#478 id=478 data-nosnippet>478</a>
-<a href=#479 id=479 data-nosnippet>479</a><span class="kw">struct
</span>EqDelRecordBatchPartnerAccessor;
-<a href=#480 id=480 data-nosnippet>480</a>
-<a href=#481 id=481 data-nosnippet>481</a><span class="kw">impl
</span>PartnerAccessor<ArrayRef> <span class="kw">for
</span>EqDelRecordBatchPartnerAccessor {
-<a href=#482 id=482 data-nosnippet>482</a> <span class="kw">fn
</span>struct_partner<<span class="lifetime">'a</span>>(<span
class="kw-2">&</span><span class="self">self</span>, schema_partner: <span
class="kw-2">&</span><span class="lifetime">'a </span>ArrayRef) -> <span
class="prelude-ty">Result</span><<span class="kw-2">&</span><span
class="lifetime">'a </span>ArrayRef> {
-<a href=#483 id=483 data-nosnippet>483</a> <span
class="prelude-val">Ok</span>(schema_partner)
-<a href=#484 id=484 data-nosnippet>484</a> }
-<a href=#485 id=485 data-nosnippet>485</a>
-<a href=#486 id=486 data-nosnippet>486</a> <span class="kw">fn
</span>field_partner<<span class="lifetime">'a</span>>(
-<a href=#487 id=487 data-nosnippet>487</a> <span
class="kw-2">&</span><span class="self">self</span>,
-<a href=#488 id=488 data-nosnippet>488</a> struct_partner: <span
class="kw-2">&</span><span class="lifetime">'a </span>ArrayRef,
-<a href=#489 id=489 data-nosnippet>489</a> field: <span
class="kw-2">&</span>NestedField,
-<a href=#490 id=490 data-nosnippet>490</a> ) -> <span
class="prelude-ty">Result</span><<span class="kw-2">&</span><span
class="lifetime">'a </span>ArrayRef> {
-<a href=#491 id=491 data-nosnippet>491</a> <span class="kw">let
</span><span class="prelude-val">Some</span>(struct_array) =
struct_partner.as_any().downcast_ref::<StructArray>() <span
class="kw">else </span>{
-<a href=#492 id=492 data-nosnippet>492</a> <span class="kw">return
</span><span class="prelude-val">Err</span>(Error::new(
-<a href=#493 id=493 data-nosnippet>493</a>
ErrorKind::Unexpected,
-<a href=#494 id=494 data-nosnippet>494</a> <span
class="string">"Expected struct array for field extraction"</span>,
-<a href=#495 id=495 data-nosnippet>495</a> ));
-<a href=#496 id=496 data-nosnippet>496</a> };
-<a href=#497 id=497 data-nosnippet>497</a>
-<a href=#498 id=498 data-nosnippet>498</a> <span class="comment">//
Find the field by name within the struct
-<a href=#499 id=499 data-nosnippet>499</a> </span><span class="kw">for
</span>(i, field_def) <span class="kw">in
</span>struct_array.fields().iter().enumerate() {
-<a href=#500 id=500 data-nosnippet>500</a> <span class="kw">if
</span>field_def.name() == <span class="kw-2">&</span>field.name {
-<a href=#501 id=501 data-nosnippet>501</a> <span
class="kw">return </span><span
class="prelude-val">Ok</span>(struct_array.column(i));
-<a href=#502 id=502 data-nosnippet>502</a> }
-<a href=#503 id=503 data-nosnippet>503</a> }
-<a href=#504 id=504 data-nosnippet>504</a>
-<a href=#505 id=505 data-nosnippet>505</a> <span
class="prelude-val">Err</span>(Error::new(
-<a href=#506 id=506 data-nosnippet>506</a> ErrorKind::Unexpected,
-<a href=#507 id=507 data-nosnippet>507</a> <span
class="macro">format!</span>(<span class="string">"Field {} not found in parent
struct"</span>, field.name),
-<a href=#508 id=508 data-nosnippet>508</a> ))
-<a href=#509 id=509 data-nosnippet>509</a> }
+<a href=#433 id=433 data-nosnippet>433</a> <span
class="prelude-val">Ok</span>((datum_iterator, field_name))
+<a href=#434 id=434 data-nosnippet>434</a> })
+<a href=#435 id=435 data-nosnippet>435</a> .collect::<<span
class="prelude-ty">Result</span><Vec<<span
class="kw">_</span>>>>()
+<a href=#436 id=436 data-nosnippet>436</a> }
+<a href=#437 id=437 data-nosnippet>437</a>}
+<a href=#438 id=438 data-nosnippet>438</a>
+<a href=#439 id=439 data-nosnippet>439</a><span class="kw">impl
</span>SchemaWithPartnerVisitor<ArrayRef> <span class="kw">for
</span>EqDelColumnProcessor<<span class="lifetime">'_</span>> {
+<a href=#440 id=440 data-nosnippet>440</a> <span class="kw">type </span>T =
();
+<a href=#441 id=441 data-nosnippet>441</a>
+<a href=#442 id=442 data-nosnippet>442</a> <span class="kw">fn
</span>schema(<span class="kw-2">&mut </span><span
class="self">self</span>, _schema: <span class="kw-2">&</span>Schema,
_partner: <span class="kw-2">&</span>ArrayRef, _value: ()) -> <span
class="prelude-ty">Result</span><()> {
+<a href=#443 id=443 data-nosnippet>443</a> <span
class="prelude-val">Ok</span>(())
+<a href=#444 id=444 data-nosnippet>444</a> }
+<a href=#445 id=445 data-nosnippet>445</a>
+<a href=#446 id=446 data-nosnippet>446</a> <span class="kw">fn
</span>field(<span class="kw-2">&mut </span><span class="self">self</span>,
field: <span class="kw-2">&</span>NestedFieldRef, partner: <span
class="kw-2">&</span>ArrayRef, _value: ()) -> <span
class="prelude-ty">Result</span><()> {
+<a href=#447 id=447 data-nosnippet>447</a> <span class="kw">if
</span><span class="self">self</span>.equality_ids.contains(<span
class="kw-2">&</span>field.id) &&
field.field_type.as_primitive_type().is_some() {
+<a href=#448 id=448 data-nosnippet>448</a> <span
class="self">self</span>.collected_columns.push((
+<a href=#449 id=449 data-nosnippet>449</a> partner.clone(),
+<a href=#450 id=450 data-nosnippet>450</a> field.name.clone(),
+<a href=#451 id=451 data-nosnippet>451</a>
field.field_type.as_ref().clone(),
+<a href=#452 id=452 data-nosnippet>452</a> ));
+<a href=#453 id=453 data-nosnippet>453</a> }
+<a href=#454 id=454 data-nosnippet>454</a> <span
class="prelude-val">Ok</span>(())
+<a href=#455 id=455 data-nosnippet>455</a> }
+<a href=#456 id=456 data-nosnippet>456</a>
+<a href=#457 id=457 data-nosnippet>457</a> <span class="kw">fn
</span>r#struct(
+<a href=#458 id=458 data-nosnippet>458</a> <span class="kw-2">&mut
</span><span class="self">self</span>,
+<a href=#459 id=459 data-nosnippet>459</a> _struct: <span
class="kw-2">&</span>StructType,
+<a href=#460 id=460 data-nosnippet>460</a> _partner: <span
class="kw-2">&</span>ArrayRef,
+<a href=#461 id=461 data-nosnippet>461</a> _results: Vec<()>,
+<a href=#462 id=462 data-nosnippet>462</a> ) -> <span
class="prelude-ty">Result</span><()> {
+<a href=#463 id=463 data-nosnippet>463</a> <span
class="prelude-val">Ok</span>(())
+<a href=#464 id=464 data-nosnippet>464</a> }
+<a href=#465 id=465 data-nosnippet>465</a>
+<a href=#466 id=466 data-nosnippet>466</a> <span class="kw">fn
</span>list(<span class="kw-2">&mut </span><span class="self">self</span>,
_list: <span class="kw-2">&</span>ListType, _partner: <span
class="kw-2">&</span>ArrayRef, _value: ()) -> <span
class="prelude-ty">Result</span><()> {
+<a href=#467 id=467 data-nosnippet>467</a> <span
class="prelude-val">Ok</span>(())
+<a href=#468 id=468 data-nosnippet>468</a> }
+<a href=#469 id=469 data-nosnippet>469</a>
+<a href=#470 id=470 data-nosnippet>470</a> <span class="kw">fn </span>map(
+<a href=#471 id=471 data-nosnippet>471</a> <span class="kw-2">&mut
</span><span class="self">self</span>,
+<a href=#472 id=472 data-nosnippet>472</a> _map: <span
class="kw-2">&</span>MapType,
+<a href=#473 id=473 data-nosnippet>473</a> _partner: <span
class="kw-2">&</span>ArrayRef,
+<a href=#474 id=474 data-nosnippet>474</a> _key_value: (),
+<a href=#475 id=475 data-nosnippet>475</a> _value: (),
+<a href=#476 id=476 data-nosnippet>476</a> ) -> <span
class="prelude-ty">Result</span><()> {
+<a href=#477 id=477 data-nosnippet>477</a> <span
class="prelude-val">Ok</span>(())
+<a href=#478 id=478 data-nosnippet>478</a> }
+<a href=#479 id=479 data-nosnippet>479</a>
+<a href=#480 id=480 data-nosnippet>480</a> <span class="kw">fn
</span>primitive(<span class="kw-2">&mut </span><span
class="self">self</span>, _primitive: <span
class="kw-2">&</span>PrimitiveType, _partner: <span
class="kw-2">&</span>ArrayRef) -> <span
class="prelude-ty">Result</span><()> {
+<a href=#481 id=481 data-nosnippet>481</a> <span
class="prelude-val">Ok</span>(())
+<a href=#482 id=482 data-nosnippet>482</a> }
+<a href=#483 id=483 data-nosnippet>483</a>}
+<a href=#484 id=484 data-nosnippet>484</a>
+<a href=#485 id=485 data-nosnippet>485</a><span class="kw">struct
</span>EqDelRecordBatchPartnerAccessor;
+<a href=#486 id=486 data-nosnippet>486</a>
+<a href=#487 id=487 data-nosnippet>487</a><span class="kw">impl
</span>PartnerAccessor<ArrayRef> <span class="kw">for
</span>EqDelRecordBatchPartnerAccessor {
+<a href=#488 id=488 data-nosnippet>488</a> <span class="kw">fn
</span>struct_partner<<span class="lifetime">'a</span>>(<span
class="kw-2">&</span><span class="self">self</span>, schema_partner: <span
class="kw-2">&</span><span class="lifetime">'a </span>ArrayRef) -> <span
class="prelude-ty">Result</span><<span class="kw-2">&</span><span
class="lifetime">'a </span>ArrayRef> {
+<a href=#489 id=489 data-nosnippet>489</a> <span
class="prelude-val">Ok</span>(schema_partner)
+<a href=#490 id=490 data-nosnippet>490</a> }
+<a href=#491 id=491 data-nosnippet>491</a>
+<a href=#492 id=492 data-nosnippet>492</a> <span class="kw">fn
</span>field_partner<<span class="lifetime">'a</span>>(
+<a href=#493 id=493 data-nosnippet>493</a> <span
class="kw-2">&</span><span class="self">self</span>,
+<a href=#494 id=494 data-nosnippet>494</a> struct_partner: <span
class="kw-2">&</span><span class="lifetime">'a </span>ArrayRef,
+<a href=#495 id=495 data-nosnippet>495</a> field: <span
class="kw-2">&</span>NestedField,
+<a href=#496 id=496 data-nosnippet>496</a> ) -> <span
class="prelude-ty">Result</span><<span class="kw-2">&</span><span
class="lifetime">'a </span>ArrayRef> {
+<a href=#497 id=497 data-nosnippet>497</a> <span class="kw">let
</span><span class="prelude-val">Some</span>(struct_array) =
struct_partner.as_any().downcast_ref::<StructArray>() <span
class="kw">else </span>{
+<a href=#498 id=498 data-nosnippet>498</a> <span class="kw">return
</span><span class="prelude-val">Err</span>(Error::new(
+<a href=#499 id=499 data-nosnippet>499</a>
ErrorKind::Unexpected,
+<a href=#500 id=500 data-nosnippet>500</a> <span
class="string">"Expected struct array for field extraction"</span>,
+<a href=#501 id=501 data-nosnippet>501</a> ));
+<a href=#502 id=502 data-nosnippet>502</a> };
+<a href=#503 id=503 data-nosnippet>503</a>
+<a href=#504 id=504 data-nosnippet>504</a> <span class="comment">//
Find the field by name within the struct
+<a href=#505 id=505 data-nosnippet>505</a> </span><span class="kw">for
</span>(i, field_def) <span class="kw">in
</span>struct_array.fields().iter().enumerate() {
+<a href=#506 id=506 data-nosnippet>506</a> <span class="kw">if
</span>field_def.name() == <span class="kw-2">&</span>field.name {
+<a href=#507 id=507 data-nosnippet>507</a> <span
class="kw">return </span><span
class="prelude-val">Ok</span>(struct_array.column(i));
+<a href=#508 id=508 data-nosnippet>508</a> }
+<a href=#509 id=509 data-nosnippet>509</a> }
<a href=#510 id=510 data-nosnippet>510</a>
-<a href=#511 id=511 data-nosnippet>511</a> <span class="kw">fn
</span>list_element_partner<<span class="lifetime">'a</span>>(<span
class="kw-2">&</span><span class="self">self</span>, _list_partner: <span
class="kw-2">&</span><span class="lifetime">'a </span>ArrayRef) -> <span
class="prelude-ty">Result</span><<span class="kw-2">&</span><span
class="lifetime">'a </span>ArrayRef> {
-<a href=#512 id=512 data-nosnippet>512</a> <span
class="prelude-val">Err</span>(Error::new(
-<a href=#513 id=513 data-nosnippet>513</a>
ErrorKind::FeatureUnsupported,
-<a href=#514 id=514 data-nosnippet>514</a> <span
class="string">"List columns are unsupported in equality deletes"</span>,
-<a href=#515 id=515 data-nosnippet>515</a> ))
-<a href=#516 id=516 data-nosnippet>516</a> }
-<a href=#517 id=517 data-nosnippet>517</a>
-<a href=#518 id=518 data-nosnippet>518</a> <span class="kw">fn
</span>map_key_partner<<span class="lifetime">'a</span>>(<span
class="kw-2">&</span><span class="self">self</span>, _map_partner: <span
class="kw-2">&</span><span class="lifetime">'a </span>ArrayRef) -> <span
class="prelude-ty">Result</span><<span class="kw-2">&</span><span
class="lifetime">'a </span>ArrayRef> {
-<a href=#519 id=519 data-nosnippet>519</a> <span
class="prelude-val">Err</span>(Error::new(
-<a href=#520 id=520 data-nosnippet>520</a>
ErrorKind::FeatureUnsupported,
-<a href=#521 id=521 data-nosnippet>521</a> <span
class="string">"Map columns are unsupported in equality deletes"</span>,
-<a href=#522 id=522 data-nosnippet>522</a> ))
-<a href=#523 id=523 data-nosnippet>523</a> }
-<a href=#524 id=524 data-nosnippet>524</a>
-<a href=#525 id=525 data-nosnippet>525</a> <span class="kw">fn
</span>map_value_partner<<span class="lifetime">'a</span>>(<span
class="kw-2">&</span><span class="self">self</span>, _map_partner: <span
class="kw-2">&</span><span class="lifetime">'a </span>ArrayRef) -> <span
class="prelude-ty">Result</span><<span class="kw-2">&</span><span
class="lifetime">'a </span>ArrayRef> {
-<a href=#526 id=526 data-nosnippet>526</a> <span
class="prelude-val">Err</span>(Error::new(
-<a href=#527 id=527 data-nosnippet>527</a>
ErrorKind::FeatureUnsupported,
-<a href=#528 id=528 data-nosnippet>528</a> <span
class="string">"Map columns are unsupported in equality deletes"</span>,
-<a href=#529 id=529 data-nosnippet>529</a> ))
-<a href=#530 id=530 data-nosnippet>530</a> }
-<a href=#531 id=531 data-nosnippet>531</a>}
-<a href=#532 id=532 data-nosnippet>532</a>
-<a href=#533 id=533 data-nosnippet>533</a><span class="attr">#[cfg(test)]
-<a href=#534 id=534 data-nosnippet>534</a></span><span class="kw">mod
</span>tests {
-<a href=#535 id=535 data-nosnippet>535</a> <span class="kw">use
</span>std::collections::HashMap;
-<a href=#536 id=536 data-nosnippet>536</a> <span class="kw">use
</span>std::fs::File;
-<a href=#537 id=537 data-nosnippet>537</a> <span class="kw">use
</span>std::sync::Arc;
+<a href=#511 id=511 data-nosnippet>511</a> <span
class="prelude-val">Err</span>(Error::new(
+<a href=#512 id=512 data-nosnippet>512</a> ErrorKind::Unexpected,
+<a href=#513 id=513 data-nosnippet>513</a> <span
class="macro">format!</span>(<span class="string">"Field {} not found in parent
struct"</span>, field.name),
+<a href=#514 id=514 data-nosnippet>514</a> ))
+<a href=#515 id=515 data-nosnippet>515</a> }
+<a href=#516 id=516 data-nosnippet>516</a>
+<a href=#517 id=517 data-nosnippet>517</a> <span class="kw">fn
</span>list_element_partner<<span class="lifetime">'a</span>>(<span
class="kw-2">&</span><span class="self">self</span>, _list_partner: <span
class="kw-2">&</span><span class="lifetime">'a </span>ArrayRef) -> <span
class="prelude-ty">Result</span><<span class="kw-2">&</span><span
class="lifetime">'a </span>ArrayRef> {
+<a href=#518 id=518 data-nosnippet>518</a> <span
class="prelude-val">Err</span>(Error::new(
+<a href=#519 id=519 data-nosnippet>519</a>
ErrorKind::FeatureUnsupported,
+<a href=#520 id=520 data-nosnippet>520</a> <span
class="string">"List columns are unsupported in equality deletes"</span>,
+<a href=#521 id=521 data-nosnippet>521</a> ))
+<a href=#522 id=522 data-nosnippet>522</a> }
+<a href=#523 id=523 data-nosnippet>523</a>
+<a href=#524 id=524 data-nosnippet>524</a> <span class="kw">fn
</span>map_key_partner<<span class="lifetime">'a</span>>(<span
class="kw-2">&</span><span class="self">self</span>, _map_partner: <span
class="kw-2">&</span><span class="lifetime">'a </span>ArrayRef) -> <span
class="prelude-ty">Result</span><<span class="kw-2">&</span><span
class="lifetime">'a </span>ArrayRef> {
+<a href=#525 id=525 data-nosnippet>525</a> <span
class="prelude-val">Err</span>(Error::new(
+<a href=#526 id=526 data-nosnippet>526</a>
ErrorKind::FeatureUnsupported,
+<a href=#527 id=527 data-nosnippet>527</a> <span
class="string">"Map columns are unsupported in equality deletes"</span>,
+<a href=#528 id=528 data-nosnippet>528</a> ))
+<a href=#529 id=529 data-nosnippet>529</a> }
+<a href=#530 id=530 data-nosnippet>530</a>
+<a href=#531 id=531 data-nosnippet>531</a> <span class="kw">fn
</span>map_value_partner<<span class="lifetime">'a</span>>(<span
class="kw-2">&</span><span class="self">self</span>, _map_partner: <span
class="kw-2">&</span><span class="lifetime">'a </span>ArrayRef) -> <span
class="prelude-ty">Result</span><<span class="kw-2">&</span><span
class="lifetime">'a </span>ArrayRef> {
+<a href=#532 id=532 data-nosnippet>532</a> <span
class="prelude-val">Err</span>(Error::new(
+<a href=#533 id=533 data-nosnippet>533</a>
ErrorKind::FeatureUnsupported,
+<a href=#534 id=534 data-nosnippet>534</a> <span
class="string">"Map columns are unsupported in equality deletes"</span>,
+<a href=#535 id=535 data-nosnippet>535</a> ))
+<a href=#536 id=536 data-nosnippet>536</a> }
+<a href=#537 id=537 data-nosnippet>537</a>}
<a href=#538 id=538 data-nosnippet>538</a>
-<a href=#539 id=539 data-nosnippet>539</a> <span class="kw">use
</span>arrow_array::{ArrayRef, Int32Array, Int64Array, RecordBatch,
StringArray, StructArray};
-<a href=#540 id=540 data-nosnippet>540</a> <span class="kw">use
</span>arrow_schema::{DataType, Field, Fields};
-<a href=#541 id=541 data-nosnippet>541</a> <span class="kw">use
</span>parquet::arrow::{ArrowWriter, PARQUET_FIELD_ID_META_KEY};
-<a href=#542 id=542 data-nosnippet>542</a> <span class="kw">use
</span>parquet::basic::Compression;
-<a href=#543 id=543 data-nosnippet>543</a> <span class="kw">use
</span>parquet::file::properties::WriterProperties;
-<a href=#544 id=544 data-nosnippet>544</a> <span class="kw">use
</span>tempfile::TempDir;
-<a href=#545 id=545 data-nosnippet>545</a>
-<a href=#546 id=546 data-nosnippet>546</a> <span class="kw">use
super</span>::<span class="kw-2">*</span>;
-<a href=#547 id=547 data-nosnippet>547</a> <span class="kw">use
</span><span class="kw">crate</span>::arrow::delete_filter::tests::setup;
-<a href=#548 id=548 data-nosnippet>548</a>
-<a href=#549 id=549 data-nosnippet>549</a> <span class="attr">#[tokio::test]
-<a href=#550 id=550 data-nosnippet>550</a> </span><span class="kw">async fn
</span>test_delete_file_loader_parse_equality_deletes() {
-<a href=#551 id=551 data-nosnippet>551</a> <span class="kw">let
</span>tmp_dir = TempDir::new().unwrap();
-<a href=#552 id=552 data-nosnippet>552</a> <span class="kw">let
</span>table_location = tmp_dir.path().as_os_str().to_str().unwrap();
-<a href=#553 id=553 data-nosnippet>553</a> <span class="kw">let
</span>file_io = FileIO::from_path(table_location).unwrap().build().unwrap();
-<a href=#554 id=554 data-nosnippet>554</a>
-<a href=#555 id=555 data-nosnippet>555</a> <span class="kw">let
</span>eq_delete_file_path = setup_write_equality_delete_file_1(table_location);
-<a href=#556 id=556 data-nosnippet>556</a>
-<a href=#557 id=557 data-nosnippet>557</a> <span class="kw">let
</span>basic_delete_file_loader = BasicDeleteFileLoader::new(file_io.clone());
-<a href=#558 id=558 data-nosnippet>558</a> <span class="kw">let
</span>record_batch_stream = basic_delete_file_loader
-<a href=#559 id=559 data-nosnippet>559</a>
.parquet_to_batch_stream(<span class="kw-2">&</span>eq_delete_file_path)
-<a href=#560 id=560 data-nosnippet>560</a> .<span class="kw">await
-<a href=#561 id=561 data-nosnippet>561</a> </span>.expect(<span
class="string">"could not get batch stream"</span>);
-<a href=#562 id=562 data-nosnippet>562</a>
-<a href=#563 id=563 data-nosnippet>563</a> <span class="kw">let
</span>eq_ids = HashSet::from_iter(<span class="macro">vec!</span>[<span
class="number">2</span>, <span class="number">3</span>, <span
class="number">4</span>, <span class="number">6</span>]);
-<a href=#564 id=564 data-nosnippet>564</a>
-<a href=#565 id=565 data-nosnippet>565</a> <span class="kw">let
</span>parsed_eq_delete =
CachingDeleteFileLoader::parse_equality_deletes_record_batch_stream(
-<a href=#566 id=566 data-nosnippet>566</a> record_batch_stream,
-<a href=#567 id=567 data-nosnippet>567</a> eq_ids,
-<a href=#568 id=568 data-nosnippet>568</a> )
-<a href=#569 id=569 data-nosnippet>569</a> .<span class="kw">await
-<a href=#570 id=570 data-nosnippet>570</a> </span>.expect(<span
class="string">"error parsing batch stream"</span>);
-<a href=#571 id=571 data-nosnippet>571</a> <span
class="macro">println!</span>(<span class="string">"{parsed_eq_delete}"</span>);
-<a href=#572 id=572 data-nosnippet>572</a>
-<a href=#573 id=573 data-nosnippet>573</a> <span class="kw">let
</span>expected = <span class="string">"((((y != 1) OR (z != 100)) OR (a !=
\"HELP\")) OR (sa != 4)) AND ((((y != 2) OR (z IS NOT NULL)) OR (a IS NOT
NULL)) OR (sa != 5))"</span>.to_string();
-<a href=#574 id=574 data-nosnippet>574</a>
-<a href=#575 id=575 data-nosnippet>575</a> <span
class="macro">assert_eq!</span>(parsed_eq_delete.to_string(), expected);
-<a href=#576 id=576 data-nosnippet>576</a> }
-<a href=#577 id=577 data-nosnippet>577</a>
-<a href=#578 id=578 data-nosnippet>578</a> <span class="doccomment">///
Create a simple field with metadata.
-<a href=#579 id=579 data-nosnippet>579</a> </span><span class="kw">fn
</span>simple_field(name: <span class="kw-2">&</span>str, ty: DataType,
nullable: bool, value: <span class="kw-2">&</span>str) -> Field {
-<a href=#580 id=580 data-nosnippet>580</a>
arrow_schema::Field::new(name, ty, nullable).with_metadata(HashMap::from([(
-<a href=#581 id=581 data-nosnippet>581</a>
PARQUET_FIELD_ID_META_KEY.to_string(),
-<a href=#582 id=582 data-nosnippet>582</a> value.to_string(),
-<a href=#583 id=583 data-nosnippet>583</a> )]))
-<a href=#584 id=584 data-nosnippet>584</a> }
-<a href=#585 id=585 data-nosnippet>585</a>
-<a href=#586 id=586 data-nosnippet>586</a> <span class="kw">fn
</span>setup_write_equality_delete_file_1(table_location: <span
class="kw-2">&</span>str) -> String {
-<a href=#587 id=587 data-nosnippet>587</a> <span class="kw">let
</span>col_y_vals = <span class="macro">vec!</span>[<span
class="number">1</span>, <span class="number">2</span>];
-<a href=#588 id=588 data-nosnippet>588</a> <span class="kw">let
</span>col_y = Arc::new(Int64Array::from(col_y_vals)) <span class="kw">as
</span>ArrayRef;
-<a href=#589 id=589 data-nosnippet>589</a>
-<a href=#590 id=590 data-nosnippet>590</a> <span class="kw">let
</span>col_z_vals = <span class="macro">vec!</span>[<span
class="prelude-val">Some</span>(<span class="number">100</span>), <span
class="prelude-val">None</span>];
-<a href=#591 id=591 data-nosnippet>591</a> <span class="kw">let
</span>col_z = Arc::new(Int64Array::from(col_z_vals)) <span class="kw">as
</span>ArrayRef;
+<a href=#539 id=539 data-nosnippet>539</a><span class="attr">#[cfg(test)]
+<a href=#540 id=540 data-nosnippet>540</a></span><span class="kw">mod
</span>tests {
+<a href=#541 id=541 data-nosnippet>541</a> <span class="kw">use
</span>std::collections::HashMap;
+<a href=#542 id=542 data-nosnippet>542</a> <span class="kw">use
</span>std::fs::File;
+<a href=#543 id=543 data-nosnippet>543</a> <span class="kw">use
</span>std::sync::Arc;
+<a href=#544 id=544 data-nosnippet>544</a>
+<a href=#545 id=545 data-nosnippet>545</a> <span class="kw">use
</span>arrow_array::cast::AsArray;
+<a href=#546 id=546 data-nosnippet>546</a> <span class="kw">use
</span>arrow_array::{ArrayRef, Int32Array, Int64Array, RecordBatch,
StringArray, StructArray};
+<a href=#547 id=547 data-nosnippet>547</a> <span class="kw">use
</span>arrow_schema::{DataType, Field, Fields};
+<a href=#548 id=548 data-nosnippet>548</a> <span class="kw">use
</span>parquet::arrow::{ArrowWriter, PARQUET_FIELD_ID_META_KEY};
+<a href=#549 id=549 data-nosnippet>549</a> <span class="kw">use
</span>parquet::basic::Compression;
+<a href=#550 id=550 data-nosnippet>550</a> <span class="kw">use
</span>parquet::file::properties::WriterProperties;
+<a href=#551 id=551 data-nosnippet>551</a> <span class="kw">use
</span>tempfile::TempDir;
+<a href=#552 id=552 data-nosnippet>552</a>
+<a href=#553 id=553 data-nosnippet>553</a> <span class="kw">use
super</span>::<span class="kw-2">*</span>;
+<a href=#554 id=554 data-nosnippet>554</a> <span class="kw">use
</span><span class="kw">crate</span>::arrow::delete_filter::tests::setup;
+<a href=#555 id=555 data-nosnippet>555</a>
+<a href=#556 id=556 data-nosnippet>556</a> <span class="attr">#[tokio::test]
+<a href=#557 id=557 data-nosnippet>557</a> </span><span class="kw">async fn
</span>test_delete_file_loader_parse_equality_deletes() {
+<a href=#558 id=558 data-nosnippet>558</a> <span class="kw">let
</span>tmp_dir = TempDir::new().unwrap();
+<a href=#559 id=559 data-nosnippet>559</a> <span class="kw">let
</span>table_location = tmp_dir.path().as_os_str().to_str().unwrap();
+<a href=#560 id=560 data-nosnippet>560</a> <span class="kw">let
</span>file_io = FileIO::from_path(table_location).unwrap().build().unwrap();
+<a href=#561 id=561 data-nosnippet>561</a>
+<a href=#562 id=562 data-nosnippet>562</a> <span class="kw">let
</span>eq_delete_file_path = setup_write_equality_delete_file_1(table_location);
+<a href=#563 id=563 data-nosnippet>563</a>
+<a href=#564 id=564 data-nosnippet>564</a> <span class="kw">let
</span>basic_delete_file_loader = BasicDeleteFileLoader::new(file_io.clone());
+<a href=#565 id=565 data-nosnippet>565</a> <span class="kw">let
</span>record_batch_stream = basic_delete_file_loader
+<a href=#566 id=566 data-nosnippet>566</a>
.parquet_to_batch_stream(<span class="kw-2">&</span>eq_delete_file_path)
+<a href=#567 id=567 data-nosnippet>567</a> .<span class="kw">await
+<a href=#568 id=568 data-nosnippet>568</a> </span>.expect(<span
class="string">"could not get batch stream"</span>);
+<a href=#569 id=569 data-nosnippet>569</a>
+<a href=#570 id=570 data-nosnippet>570</a> <span class="kw">let
</span>eq_ids = HashSet::from_iter(<span class="macro">vec!</span>[<span
class="number">2</span>, <span class="number">3</span>, <span
class="number">4</span>, <span class="number">6</span>]);
+<a href=#571 id=571 data-nosnippet>571</a>
+<a href=#572 id=572 data-nosnippet>572</a> <span class="kw">let
</span>parsed_eq_delete =
CachingDeleteFileLoader::parse_equality_deletes_record_batch_stream(
+<a href=#573 id=573 data-nosnippet>573</a> record_batch_stream,
+<a href=#574 id=574 data-nosnippet>574</a> eq_ids,
+<a href=#575 id=575 data-nosnippet>575</a> )
+<a href=#576 id=576 data-nosnippet>576</a> .<span class="kw">await
+<a href=#577 id=577 data-nosnippet>577</a> </span>.expect(<span
class="string">"error parsing batch stream"</span>);
+<a href=#578 id=578 data-nosnippet>578</a> <span
class="macro">println!</span>(<span class="string">"{parsed_eq_delete}"</span>);
+<a href=#579 id=579 data-nosnippet>579</a>
+<a href=#580 id=580 data-nosnippet>580</a> <span class="kw">let
</span>expected = <span class="string">"((((y != 1) OR (z != 100)) OR (a !=
\"HELP\")) OR (sa != 4)) AND ((((y != 2) OR (z IS NOT NULL)) OR (a IS NOT
NULL)) OR (sa != 5))"</span>.to_string();
+<a href=#581 id=581 data-nosnippet>581</a>
+<a href=#582 id=582 data-nosnippet>582</a> <span
class="macro">assert_eq!</span>(parsed_eq_delete.to_string(), expected);
+<a href=#583 id=583 data-nosnippet>583</a> }
+<a href=#584 id=584 data-nosnippet>584</a>
+<a href=#585 id=585 data-nosnippet>585</a> <span class="doccomment">///
Create a simple field with metadata.
+<a href=#586 id=586 data-nosnippet>586</a> </span><span class="kw">fn
</span>simple_field(name: <span class="kw-2">&</span>str, ty: DataType,
nullable: bool, value: <span class="kw-2">&</span>str) -> Field {
+<a href=#587 id=587 data-nosnippet>587</a>
arrow_schema::Field::new(name, ty, nullable).with_metadata(HashMap::from([(
+<a href=#588 id=588 data-nosnippet>588</a>
PARQUET_FIELD_ID_META_KEY.to_string(),
+<a href=#589 id=589 data-nosnippet>589</a> value.to_string(),
+<a href=#590 id=590 data-nosnippet>590</a> )]))
+<a href=#591 id=591 data-nosnippet>591</a> }
<a href=#592 id=592 data-nosnippet>592</a>
-<a href=#593 id=593 data-nosnippet>593</a> <span class="kw">let
</span>col_a_vals = <span class="macro">vec!</span>[<span
class="prelude-val">Some</span>(<span class="string">"HELP"</span>), <span
class="prelude-val">None</span>];
-<a href=#594 id=594 data-nosnippet>594</a> <span class="kw">let
</span>col_a = Arc::new(StringArray::from(col_a_vals)) <span class="kw">as
</span>ArrayRef;
-<a href=#595 id=595 data-nosnippet>595</a>
-<a href=#596 id=596 data-nosnippet>596</a> <span class="kw">let
</span>col_s = Arc::new(StructArray::from(<span class="macro">vec!</span>[
-<a href=#597 id=597 data-nosnippet>597</a> (
-<a href=#598 id=598 data-nosnippet>598</a>
Arc::new(simple_field(<span class="string">"sa"</span>, DataType::Int32, <span
class="bool-val">false</span>, <span class="string">"6"</span>)),
-<a href=#599 id=599 data-nosnippet>599</a>
Arc::new(Int32Array::from(<span class="macro">vec!</span>[<span
class="number">4</span>, <span class="number">5</span>])) <span class="kw">as
</span>ArrayRef,
-<a href=#600 id=600 data-nosnippet>600</a> ),
-<a href=#601 id=601 data-nosnippet>601</a> (
-<a href=#602 id=602 data-nosnippet>602</a>
Arc::new(simple_field(<span class="string">"sb"</span>, DataType::Utf8, <span
class="bool-val">true</span>, <span class="string">"7"</span>)),
-<a href=#603 id=603 data-nosnippet>603</a>
Arc::new(StringArray::from(<span class="macro">vec!</span>[<span
class="prelude-val">Some</span>(<span class="string">"x"</span>), <span
class="prelude-val">None</span>])) <span class="kw">as </span>ArrayRef,
-<a href=#604 id=604 data-nosnippet>604</a> ),
-<a href=#605 id=605 data-nosnippet>605</a> ]));
-<a href=#606 id=606 data-nosnippet>606</a>
-<a href=#607 id=607 data-nosnippet>607</a> <span class="kw">let
</span>equality_delete_schema = {
-<a href=#608 id=608 data-nosnippet>608</a> <span class="kw">let
</span>struct_field = DataType::Struct(Fields::from(<span
class="macro">vec!</span>[
-<a href=#609 id=609 data-nosnippet>609</a> simple_field(<span
class="string">"sa"</span>, DataType::Int32, <span
class="bool-val">false</span>, <span class="string">"6"</span>),
-<a href=#610 id=610 data-nosnippet>610</a> simple_field(<span
class="string">"sb"</span>, DataType::Utf8, <span class="bool-val">true</span>,
<span class="string">"7"</span>),
-<a href=#611 id=611 data-nosnippet>611</a> ]));
-<a href=#612 id=612 data-nosnippet>612</a>
-<a href=#613 id=613 data-nosnippet>613</a> <span class="kw">let
</span>fields = <span class="macro">vec!</span>[
-<a href=#614 id=614 data-nosnippet>614</a> Field::new(<span
class="string">"y"</span>, arrow_schema::DataType::Int64, <span
class="bool-val">true</span>).with_metadata(HashMap::from(
-<a href=#615 id=615 data-nosnippet>615</a>
[(PARQUET_FIELD_ID_META_KEY.to_string(), <span
class="string">"2"</span>.to_string())],
-<a href=#616 id=616 data-nosnippet>616</a> )),
-<a href=#617 id=617 data-nosnippet>617</a> Field::new(<span
class="string">"z"</span>, arrow_schema::DataType::Int64, <span
class="bool-val">true</span>).with_metadata(HashMap::from(
-<a href=#618 id=618 data-nosnippet>618</a>
[(PARQUET_FIELD_ID_META_KEY.to_string(), <span
class="string">"3"</span>.to_string())],
-<a href=#619 id=619 data-nosnippet>619</a> )),
-<a href=#620 id=620 data-nosnippet>620</a> Field::new(<span
class="string">"a"</span>, arrow_schema::DataType::Utf8, <span
class="bool-val">true</span>).with_metadata(HashMap::from([
-<a href=#621 id=621 data-nosnippet>621</a>
(PARQUET_FIELD_ID_META_KEY.to_string(), <span
class="string">"4"</span>.to_string()),
-<a href=#622 id=622 data-nosnippet>622</a> ])),
-<a href=#623 id=623 data-nosnippet>623</a> simple_field(<span
class="string">"s"</span>, struct_field, <span class="bool-val">false</span>,
<span class="string">"5"</span>),
-<a href=#624 id=624 data-nosnippet>624</a> ];
-<a href=#625 id=625 data-nosnippet>625</a>
Arc::new(arrow_schema::Schema::new(fields))
-<a href=#626 id=626 data-nosnippet>626</a> };
-<a href=#627 id=627 data-nosnippet>627</a>
-<a href=#628 id=628 data-nosnippet>628</a> <span class="kw">let
</span>equality_deletes_to_write =
RecordBatch::try_new(equality_delete_schema.clone(), <span
class="macro">vec!</span>[
-<a href=#629 id=629 data-nosnippet>629</a> col_y, col_z, col_a,
col_s,
-<a href=#630 id=630 data-nosnippet>630</a> ])
-<a href=#631 id=631 data-nosnippet>631</a> .unwrap();
-<a href=#632 id=632 data-nosnippet>632</a>
-<a href=#633 id=633 data-nosnippet>633</a> <span class="kw">let
</span>path = <span class="macro">format!</span>(<span
class="string">"{}/equality-deletes-1.parquet"</span>, <span
class="kw-2">&</span>table_location);
+<a href=#593 id=593 data-nosnippet>593</a> <span class="kw">fn
</span>setup_write_equality_delete_file_1(table_location: <span
class="kw-2">&</span>str) -> String {
+<a href=#594 id=594 data-nosnippet>594</a> <span class="kw">let
</span>col_y_vals = <span class="macro">vec!</span>[<span
class="number">1</span>, <span class="number">2</span>];
+<a href=#595 id=595 data-nosnippet>595</a> <span class="kw">let
</span>col_y = Arc::new(Int64Array::from(col_y_vals)) <span class="kw">as
</span>ArrayRef;
+<a href=#596 id=596 data-nosnippet>596</a>
+<a href=#597 id=597 data-nosnippet>597</a> <span class="kw">let
</span>col_z_vals = <span class="macro">vec!</span>[<span
class="prelude-val">Some</span>(<span class="number">100</span>), <span
class="prelude-val">None</span>];
+<a href=#598 id=598 data-nosnippet>598</a> <span class="kw">let
</span>col_z = Arc::new(Int64Array::from(col_z_vals)) <span class="kw">as
</span>ArrayRef;
+<a href=#599 id=599 data-nosnippet>599</a>
+<a href=#600 id=600 data-nosnippet>600</a> <span class="kw">let
</span>col_a_vals = <span class="macro">vec!</span>[<span
class="prelude-val">Some</span>(<span class="string">"HELP"</span>), <span
class="prelude-val">None</span>];
+<a href=#601 id=601 data-nosnippet>601</a> <span class="kw">let
</span>col_a = Arc::new(StringArray::from(col_a_vals)) <span class="kw">as
</span>ArrayRef;
+<a href=#602 id=602 data-nosnippet>602</a>
+<a href=#603 id=603 data-nosnippet>603</a> <span class="kw">let
</span>col_s = Arc::new(StructArray::from(<span class="macro">vec!</span>[
+<a href=#604 id=604 data-nosnippet>604</a> (
+<a href=#605 id=605 data-nosnippet>605</a>
Arc::new(simple_field(<span class="string">"sa"</span>, DataType::Int32, <span
class="bool-val">false</span>, <span class="string">"6"</span>)),
+<a href=#606 id=606 data-nosnippet>606</a>
Arc::new(Int32Array::from(<span class="macro">vec!</span>[<span
class="number">4</span>, <span class="number">5</span>])) <span class="kw">as
</span>ArrayRef,
+<a href=#607 id=607 data-nosnippet>607</a> ),
+<a href=#608 id=608 data-nosnippet>608</a> (
+<a href=#609 id=609 data-nosnippet>609</a>
Arc::new(simple_field(<span class="string">"sb"</span>, DataType::Utf8, <span
class="bool-val">true</span>, <span class="string">"7"</span>)),
+<a href=#610 id=610 data-nosnippet>610</a>
Arc::new(StringArray::from(<span class="macro">vec!</span>[<span
class="prelude-val">Some</span>(<span class="string">"x"</span>), <span
class="prelude-val">None</span>])) <span class="kw">as </span>ArrayRef,
+<a href=#611 id=611 data-nosnippet>611</a> ),
+<a href=#612 id=612 data-nosnippet>612</a> ]));
+<a href=#613 id=613 data-nosnippet>613</a>
+<a href=#614 id=614 data-nosnippet>614</a> <span class="kw">let
</span>equality_delete_schema = {
+<a href=#615 id=615 data-nosnippet>615</a> <span class="kw">let
</span>struct_field = DataType::Struct(Fields::from(<span
class="macro">vec!</span>[
+<a href=#616 id=616 data-nosnippet>616</a> simple_field(<span
class="string">"sa"</span>, DataType::Int32, <span
class="bool-val">false</span>, <span class="string">"6"</span>),
+<a href=#617 id=617 data-nosnippet>617</a> simple_field(<span
class="string">"sb"</span>, DataType::Utf8, <span class="bool-val">true</span>,
<span class="string">"7"</span>),
+<a href=#618 id=618 data-nosnippet>618</a> ]));
+<a href=#619 id=619 data-nosnippet>619</a>
+<a href=#620 id=620 data-nosnippet>620</a> <span class="kw">let
</span>fields = <span class="macro">vec!</span>[
+<a href=#621 id=621 data-nosnippet>621</a> Field::new(<span
class="string">"y"</span>, arrow_schema::DataType::Int64, <span
class="bool-val">true</span>).with_metadata(HashMap::from(
+<a href=#622 id=622 data-nosnippet>622</a>
[(PARQUET_FIELD_ID_META_KEY.to_string(), <span
class="string">"2"</span>.to_string())],
+<a href=#623 id=623 data-nosnippet>623</a> )),
+<a href=#624 id=624 data-nosnippet>624</a> Field::new(<span
class="string">"z"</span>, arrow_schema::DataType::Int64, <span
class="bool-val">true</span>).with_metadata(HashMap::from(
+<a href=#625 id=625 data-nosnippet>625</a>
[(PARQUET_FIELD_ID_META_KEY.to_string(), <span
class="string">"3"</span>.to_string())],
+<a href=#626 id=626 data-nosnippet>626</a> )),
+<a href=#627 id=627 data-nosnippet>627</a> Field::new(<span
class="string">"a"</span>, arrow_schema::DataType::Utf8, <span
class="bool-val">true</span>).with_metadata(HashMap::from([
+<a href=#628 id=628 data-nosnippet>628</a>
(PARQUET_FIELD_ID_META_KEY.to_string(), <span
class="string">"4"</span>.to_string()),
+<a href=#629 id=629 data-nosnippet>629</a> ])),
+<a href=#630 id=630 data-nosnippet>630</a> simple_field(<span
class="string">"s"</span>, struct_field, <span class="bool-val">false</span>,
<span class="string">"5"</span>),
+<a href=#631 id=631 data-nosnippet>631</a> ];
+<a href=#632 id=632 data-nosnippet>632</a>
Arc::new(arrow_schema::Schema::new(fields))
+<a href=#633 id=633 data-nosnippet>633</a> };
<a href=#634 id=634 data-nosnippet>634</a>
-<a href=#635 id=635 data-nosnippet>635</a> <span class="kw">let
</span>file = File::create(<span class="kw-2">&</span>path).unwrap();
-<a href=#636 id=636 data-nosnippet>636</a>
-<a href=#637 id=637 data-nosnippet>637</a> <span class="kw">let
</span>props = WriterProperties::builder()
-<a href=#638 id=638 data-nosnippet>638</a>
.set_compression(Compression::SNAPPY)
-<a href=#639 id=639 data-nosnippet>639</a> .build();
-<a href=#640 id=640 data-nosnippet>640</a>
-<a href=#641 id=641 data-nosnippet>641</a> <span class="kw">let
</span><span class="kw-2">mut </span>writer = ArrowWriter::try_new(
-<a href=#642 id=642 data-nosnippet>642</a> file,
-<a href=#643 id=643 data-nosnippet>643</a>
equality_deletes_to_write.schema(),
-<a href=#644 id=644 data-nosnippet>644</a> <span
class="prelude-val">Some</span>(props.clone()),
-<a href=#645 id=645 data-nosnippet>645</a> )
-<a href=#646 id=646 data-nosnippet>646</a> .unwrap();
+<a href=#635 id=635 data-nosnippet>635</a> <span class="kw">let
</span>equality_deletes_to_write =
RecordBatch::try_new(equality_delete_schema.clone(), <span
class="macro">vec!</span>[
+<a href=#636 id=636 data-nosnippet>636</a> col_y, col_z, col_a,
col_s,
+<a href=#637 id=637 data-nosnippet>637</a> ])
+<a href=#638 id=638 data-nosnippet>638</a> .unwrap();
+<a href=#639 id=639 data-nosnippet>639</a>
+<a href=#640 id=640 data-nosnippet>640</a> <span class="kw">let
</span>path = <span class="macro">format!</span>(<span
class="string">"{}/equality-deletes-1.parquet"</span>, <span
class="kw-2">&</span>table_location);
+<a href=#641 id=641 data-nosnippet>641</a>
+<a href=#642 id=642 data-nosnippet>642</a> <span class="kw">let
</span>file = File::create(<span class="kw-2">&</span>path).unwrap();
+<a href=#643 id=643 data-nosnippet>643</a>
+<a href=#644 id=644 data-nosnippet>644</a> <span class="kw">let
</span>props = WriterProperties::builder()
+<a href=#645 id=645 data-nosnippet>645</a>
.set_compression(Compression::SNAPPY)
+<a href=#646 id=646 data-nosnippet>646</a> .build();
<a href=#647 id=647 data-nosnippet>647</a>
-<a href=#648 id=648 data-nosnippet>648</a> writer
-<a href=#649 id=649 data-nosnippet>649</a> .write(<span
class="kw-2">&</span>equality_deletes_to_write)
-<a href=#650 id=650 data-nosnippet>650</a> .expect(<span
class="string">"Writing batch"</span>);
-<a href=#651 id=651 data-nosnippet>651</a>
-<a href=#652 id=652 data-nosnippet>652</a> <span class="comment">//
writer must be closed to write footer
-<a href=#653 id=653 data-nosnippet>653</a>
</span>writer.close().unwrap();
+<a href=#648 id=648 data-nosnippet>648</a> <span class="kw">let
</span><span class="kw-2">mut </span>writer = ArrowWriter::try_new(
+<a href=#649 id=649 data-nosnippet>649</a> file,
+<a href=#650 id=650 data-nosnippet>650</a>
equality_deletes_to_write.schema(),
+<a href=#651 id=651 data-nosnippet>651</a> <span
class="prelude-val">Some</span>(props.clone()),
+<a href=#652 id=652 data-nosnippet>652</a> )
+<a href=#653 id=653 data-nosnippet>653</a> .unwrap();
<a href=#654 id=654 data-nosnippet>654</a>
-<a href=#655 id=655 data-nosnippet>655</a> path
-<a href=#656 id=656 data-nosnippet>656</a> }
-<a href=#657 id=657 data-nosnippet>657</a>
-<a href=#658 id=658 data-nosnippet>658</a> <span class="attr">#[tokio::test]
-<a href=#659 id=659 data-nosnippet>659</a> </span><span class="kw">async fn
</span>test_caching_delete_file_loader_load_deletes() {
-<a href=#660 id=660 data-nosnippet>660</a> <span class="kw">let
</span>tmp_dir = TempDir::new().unwrap();
-<a href=#661 id=661 data-nosnippet>661</a> <span class="kw">let
</span>table_location = tmp_dir.path();
-<a href=#662 id=662 data-nosnippet>662</a> <span class="kw">let
</span>file_io = FileIO::from_path(table_location.as_os_str().to_str().unwrap())
-<a href=#663 id=663 data-nosnippet>663</a> .unwrap()
-<a href=#664 id=664 data-nosnippet>664</a> .build()
-<a href=#665 id=665 data-nosnippet>665</a> .unwrap();
-<a href=#666 id=666 data-nosnippet>666</a>
-<a href=#667 id=667 data-nosnippet>667</a> <span class="kw">let
</span>delete_file_loader = CachingDeleteFileLoader::new(file_io.clone(), <span
class="number">10</span>);
-<a href=#668 id=668 data-nosnippet>668</a>
-<a href=#669 id=669 data-nosnippet>669</a> <span class="kw">let
</span>file_scan_tasks = setup(table_location);
-<a href=#670 id=670 data-nosnippet>670</a>
-<a href=#671 id=671 data-nosnippet>671</a> <span class="kw">let
</span>delete_filter = delete_file_loader
-<a href=#672 id=672 data-nosnippet>672</a> .load_deletes(<span
class="kw-2">&</span>file_scan_tasks[<span
class="number">0</span>].deletes, file_scan_tasks[<span
class="number">0</span>].schema_ref())
-<a href=#673 id=673 data-nosnippet>673</a> .<span class="kw">await
-<a href=#674 id=674 data-nosnippet>674</a> </span>.unwrap()
-<a href=#675 id=675 data-nosnippet>675</a> .unwrap();
-<a href=#676 id=676 data-nosnippet>676</a>
-<a href=#677 id=677 data-nosnippet>677</a> <span class="kw">let
</span>result = delete_filter
-<a href=#678 id=678 data-nosnippet>678</a> .get_delete_vector(<span
class="kw-2">&</span>file_scan_tasks[<span class="number">0</span>])
-<a href=#679 id=679 data-nosnippet>679</a> .unwrap();
-<a href=#680 id=680 data-nosnippet>680</a>
-<a href=#681 id=681 data-nosnippet>681</a> <span class="comment">//
union of pos dels from pos del file 1 and 2, ie
-<a href=#682 id=682 data-nosnippet>682</a> // [0, 1, 3, 5, 6, 8, 1022,
1023] | [0, 1, 3, 5, 20, 21, 22, 23]
-<a href=#683 id=683 data-nosnippet>683</a> // = [0, 1, 3, 5, 6, 8, 20,
21, 22, 23, 1022, 1023]
-<a href=#684 id=684 data-nosnippet>684</a> </span><span
class="macro">assert_eq!</span>(result.lock().unwrap().len(), <span
class="number">12</span>);
-<a href=#685 id=685 data-nosnippet>685</a>
-<a href=#686 id=686 data-nosnippet>686</a> <span class="kw">let
</span>result = delete_filter.get_delete_vector(<span
class="kw-2">&</span>file_scan_tasks[<span class="number">1</span>]);
-<a href=#687 id=687 data-nosnippet>687</a> <span
class="macro">assert!</span>(result.is_none()); <span class="comment">// no pos
dels for file 3
-<a href=#688 id=688 data-nosnippet>688</a> </span>}
-<a href=#689 id=689
data-nosnippet>689</a>}</code></pre></div></section></main></body></html>
\ No newline at end of file
+<a href=#655 id=655 data-nosnippet>655</a> writer
+<a href=#656 id=656 data-nosnippet>656</a> .write(<span
class="kw-2">&</span>equality_deletes_to_write)
+<a href=#657 id=657 data-nosnippet>657</a> .expect(<span
class="string">"Writing batch"</span>);
+<a href=#658 id=658 data-nosnippet>658</a>
+<a href=#659 id=659 data-nosnippet>659</a> <span class="comment">//
writer must be closed to write footer
+<a href=#660 id=660 data-nosnippet>660</a>
</span>writer.close().unwrap();
+<a href=#661 id=661 data-nosnippet>661</a>
+<a href=#662 id=662 data-nosnippet>662</a> path
+<a href=#663 id=663 data-nosnippet>663</a> }
+<a href=#664 id=664 data-nosnippet>664</a>
+<a href=#665 id=665 data-nosnippet>665</a> <span class="attr">#[tokio::test]
+<a href=#666 id=666 data-nosnippet>666</a> </span><span class="kw">async fn
</span>test_caching_delete_file_loader_load_deletes() {
+<a href=#667 id=667 data-nosnippet>667</a> <span class="kw">let
</span>tmp_dir = TempDir::new().unwrap();
+<a href=#668 id=668 data-nosnippet>668</a> <span class="kw">let
</span>table_location = tmp_dir.path();
+<a href=#669 id=669 data-nosnippet>669</a> <span class="kw">let
</span>file_io = FileIO::from_path(table_location.as_os_str().to_str().unwrap())
+<a href=#670 id=670 data-nosnippet>670</a> .unwrap()
+<a href=#671 id=671 data-nosnippet>671</a> .build()
+<a href=#672 id=672 data-nosnippet>672</a> .unwrap();
+<a href=#673 id=673 data-nosnippet>673</a>
+<a href=#674 id=674 data-nosnippet>674</a> <span class="kw">let
</span>delete_file_loader = CachingDeleteFileLoader::new(file_io.clone(), <span
class="number">10</span>);
+<a href=#675 id=675 data-nosnippet>675</a>
+<a href=#676 id=676 data-nosnippet>676</a> <span class="kw">let
</span>file_scan_tasks = setup(table_location);
+<a href=#677 id=677 data-nosnippet>677</a>
+<a href=#678 id=678 data-nosnippet>678</a> <span class="kw">let
</span>delete_filter = delete_file_loader
+<a href=#679 id=679 data-nosnippet>679</a> .load_deletes(<span
class="kw-2">&</span>file_scan_tasks[<span
class="number">0</span>].deletes, file_scan_tasks[<span
class="number">0</span>].schema_ref())
+<a href=#680 id=680 data-nosnippet>680</a> .<span class="kw">await
+<a href=#681 id=681 data-nosnippet>681</a> </span>.unwrap()
+<a href=#682 id=682 data-nosnippet>682</a> .unwrap();
+<a href=#683 id=683 data-nosnippet>683</a>
+<a href=#684 id=684 data-nosnippet>684</a> <span class="kw">let
</span>result = delete_filter
+<a href=#685 id=685 data-nosnippet>685</a> .get_delete_vector(<span
class="kw-2">&</span>file_scan_tasks[<span class="number">0</span>])
+<a href=#686 id=686 data-nosnippet>686</a> .unwrap();
+<a href=#687 id=687 data-nosnippet>687</a>
+<a href=#688 id=688 data-nosnippet>688</a> <span class="comment">//
union of pos dels from pos del file 1 and 2, ie
+<a href=#689 id=689 data-nosnippet>689</a> // [0, 1, 3, 5, 6, 8, 1022,
1023] | [0, 1, 3, 5, 20, 21, 22, 23]
+<a href=#690 id=690 data-nosnippet>690</a> // = [0, 1, 3, 5, 6, 8, 20,
21, 22, 23, 1022, 1023]
+<a href=#691 id=691 data-nosnippet>691</a> </span><span
class="macro">assert_eq!</span>(result.lock().unwrap().len(), <span
class="number">12</span>);
+<a href=#692 id=692 data-nosnippet>692</a>
+<a href=#693 id=693 data-nosnippet>693</a> <span class="kw">let
</span>result = delete_filter.get_delete_vector(<span
class="kw-2">&</span>file_scan_tasks[<span class="number">1</span>]);
+<a href=#694 id=694 data-nosnippet>694</a> <span
class="macro">assert!</span>(result.is_none()); <span class="comment">// no pos
dels for file 3
+<a href=#695 id=695 data-nosnippet>695</a> </span>}
+<a href=#696 id=696 data-nosnippet>696</a>
+<a href=#697 id=697 data-nosnippet>697</a> <span class="doccomment">///
Verifies that evolve_schema on partial-schema equality deletes works correctly
+<a href=#698 id=698 data-nosnippet>698</a> /// when only equality_ids
columns are evolved, not all table columns.
+<a href=#699 id=699 data-nosnippet>699</a> ///
+<a href=#700 id=700 data-nosnippet>700</a> /// Per the [Iceberg
spec](https://iceberg.apache.org/spec/#equality-delete-files),
+<a href=#701 id=701 data-nosnippet>701</a> /// equality delete files can
contain only a subset of columns.
+<a href=#702 id=702 data-nosnippet>702</a> </span><span
class="attr">#[tokio::test]
+<a href=#703 id=703 data-nosnippet>703</a> </span><span class="kw">async fn
</span>test_partial_schema_equality_deletes_evolve_succeeds() {
+<a href=#704 id=704 data-nosnippet>704</a> <span class="kw">let
</span>tmp_dir = TempDir::new().unwrap();
+<a href=#705 id=705 data-nosnippet>705</a> <span class="kw">let
</span>table_location = tmp_dir.path().as_os_str().to_str().unwrap();
+<a href=#706 id=706 data-nosnippet>706</a>
+<a href=#707 id=707 data-nosnippet>707</a> <span class="comment">//
Create table schema with REQUIRED fields
+<a href=#708 id=708 data-nosnippet>708</a> </span><span class="kw">let
</span>table_schema = Arc::new(
+<a href=#709 id=709 data-nosnippet>709</a> Schema::builder()
+<a href=#710 id=710 data-nosnippet>710</a>
.with_schema_id(<span class="number">1</span>)
+<a href=#711 id=711 data-nosnippet>711</a> .with_fields(<span
class="macro">vec!</span>[
+<a href=#712 id=712 data-nosnippet>712</a> <span
class="kw">crate</span>::spec::NestedField::required(
+<a href=#713 id=713 data-nosnippet>713</a> <span
class="number">1</span>,
+<a href=#714 id=714 data-nosnippet>714</a> <span
class="string">"id"</span>,
+<a href=#715 id=715 data-nosnippet>715</a> <span
class="kw">crate</span>::spec::Type::Primitive(<span
class="kw">crate</span>::spec::PrimitiveType::Int),
+<a href=#716 id=716 data-nosnippet>716</a> )
+<a href=#717 id=717 data-nosnippet>717</a> .into(),
+<a href=#718 id=718 data-nosnippet>718</a> <span
class="kw">crate</span>::spec::NestedField::required(
+<a href=#719 id=719 data-nosnippet>719</a> <span
class="number">2</span>,
+<a href=#720 id=720 data-nosnippet>720</a> <span
class="string">"data"</span>,
+<a href=#721 id=721 data-nosnippet>721</a> <span
class="kw">crate</span>::spec::Type::Primitive(<span
class="kw">crate</span>::spec::PrimitiveType::String),
+<a href=#722 id=722 data-nosnippet>722</a> )
+<a href=#723 id=723 data-nosnippet>723</a> .into(),
+<a href=#724 id=724 data-nosnippet>724</a> ])
+<a href=#725 id=725 data-nosnippet>725</a> .build()
+<a href=#726 id=726 data-nosnippet>726</a> .unwrap(),
+<a href=#727 id=727 data-nosnippet>727</a> );
+<a href=#728 id=728 data-nosnippet>728</a>
+<a href=#729 id=729 data-nosnippet>729</a> <span class="comment">//
Write equality delete file with PARTIAL schema (only 'data' column)
+<a href=#730 id=730 data-nosnippet>730</a> </span><span class="kw">let
</span>delete_file_path = {
+<a href=#731 id=731 data-nosnippet>731</a> <span class="kw">let
</span>data_vals = <span class="macro">vec!</span>[<span
class="string">"a"</span>, <span class="string">"d"</span>, <span
class="string">"g"</span>];
+<a href=#732 id=732 data-nosnippet>732</a> <span class="kw">let
</span>data_col = Arc::new(StringArray::from(data_vals)) <span class="kw">as
</span>ArrayRef;
+<a href=#733 id=733 data-nosnippet>733</a>
+<a href=#734 id=734 data-nosnippet>734</a> <span class="kw">let
</span>delete_schema = Arc::new(arrow_schema::Schema::new(<span
class="macro">vec!</span>[simple_field(
+<a href=#735 id=735 data-nosnippet>735</a> <span
class="string">"data"</span>,
+<a href=#736 id=736 data-nosnippet>736</a> DataType::Utf8,
+<a href=#737 id=737 data-nosnippet>737</a> <span
class="bool-val">false</span>,
+<a href=#738 id=738 data-nosnippet>738</a> <span
class="string">"2"</span>, <span class="comment">// field ID
+<a href=#739 id=739 data-nosnippet>739</a> </span>)]));
+<a href=#740 id=740 data-nosnippet>740</a>
+<a href=#741 id=741 data-nosnippet>741</a> <span class="kw">let
</span>delete_batch = RecordBatch::try_new(delete_schema.clone(), <span
class="macro">vec!</span>[data_col]).unwrap();
+<a href=#742 id=742 data-nosnippet>742</a>
+<a href=#743 id=743 data-nosnippet>743</a> <span class="kw">let
</span>path = <span class="macro">format!</span>(<span
class="string">"{}/partial-eq-deletes.parquet"</span>, <span
class="kw-2">&</span>table_location);
+<a href=#744 id=744 data-nosnippet>744</a> <span class="kw">let
</span>file = File::create(<span class="kw-2">&</span>path).unwrap();
+<a href=#745 id=745 data-nosnippet>745</a> <span class="kw">let
</span>props = WriterProperties::builder()
+<a href=#746 id=746 data-nosnippet>746</a>
.set_compression(Compression::SNAPPY)
+<a href=#747 id=747 data-nosnippet>747</a> .build();
+<a href=#748 id=748 data-nosnippet>748</a> <span class="kw">let
</span><span class="kw-2">mut </span>writer =
+<a href=#749 id=749 data-nosnippet>749</a>
ArrowWriter::try_new(file, delete_batch.schema(), <span
class="prelude-val">Some</span>(props)).unwrap();
+<a href=#750 id=750 data-nosnippet>750</a> writer.write(<span
class="kw-2">&</span>delete_batch).expect(<span class="string">"Writing
batch"</span>);
+<a href=#751 id=751 data-nosnippet>751</a> writer.close().unwrap();
+<a href=#752 id=752 data-nosnippet>752</a> path
+<a href=#753 id=753 data-nosnippet>753</a> };
+<a href=#754 id=754 data-nosnippet>754</a>
+<a href=#755 id=755 data-nosnippet>755</a> <span class="kw">let
</span>file_io = FileIO::from_path(table_location).unwrap().build().unwrap();
+<a href=#756 id=756 data-nosnippet>756</a> <span class="kw">let
</span>basic_delete_file_loader = BasicDeleteFileLoader::new(file_io.clone());
+<a href=#757 id=757 data-nosnippet>757</a>
+<a href=#758 id=758 data-nosnippet>758</a> <span class="kw">let
</span>batch_stream = basic_delete_file_loader
+<a href=#759 id=759 data-nosnippet>759</a>
.parquet_to_batch_stream(<span class="kw-2">&</span>delete_file_path)
+<a href=#760 id=760 data-nosnippet>760</a> .<span class="kw">await
+<a href=#761 id=761 data-nosnippet>761</a> </span>.unwrap();
+<a href=#762 id=762 data-nosnippet>762</a>
+<a href=#763 id=763 data-nosnippet>763</a> <span class="comment">//
Only evolve the equality_ids columns (field 2), not all table columns
+<a href=#764 id=764 data-nosnippet>764</a> </span><span class="kw">let
</span>equality_ids = <span class="macro">vec!</span>[<span
class="number">2</span>];
+<a href=#765 id=765 data-nosnippet>765</a> <span class="kw">let
</span>evolved_stream =
+<a href=#766 id=766 data-nosnippet>766</a>
BasicDeleteFileLoader::evolve_schema(batch_stream, table_schema, <span
class="kw-2">&</span>equality_ids)
+<a href=#767 id=767 data-nosnippet>767</a> .<span
class="kw">await
+<a href=#768 id=768 data-nosnippet>768</a> </span>.unwrap();
+<a href=#769 id=769 data-nosnippet>769</a>
+<a href=#770 id=770 data-nosnippet>770</a> <span class="kw">let
</span>result = evolved_stream.try_collect::<Vec<<span
class="kw">_</span>>>().<span class="kw">await</span>;
+<a href=#771 id=771 data-nosnippet>771</a>
+<a href=#772 id=772 data-nosnippet>772</a> <span
class="macro">assert!</span>(
+<a href=#773 id=773 data-nosnippet>773</a> result.is_ok(),
+<a href=#774 id=774 data-nosnippet>774</a> <span
class="string">"Expected success when evolving only equality_ids columns, got
error: {:?}"</span>,
+<a href=#775 id=775 data-nosnippet>775</a> result.err()
+<a href=#776 id=776 data-nosnippet>776</a> );
+<a href=#777 id=777 data-nosnippet>777</a>
+<a href=#778 id=778 data-nosnippet>778</a> <span class="kw">let
</span>batches = result.unwrap();
+<a href=#779 id=779 data-nosnippet>779</a> <span
class="macro">assert_eq!</span>(batches.len(), <span class="number">1</span>);
+<a href=#780 id=780 data-nosnippet>780</a>
+<a href=#781 id=781 data-nosnippet>781</a> <span class="kw">let
</span>batch = <span class="kw-2">&</span>batches[<span
class="number">0</span>];
+<a href=#782 id=782 data-nosnippet>782</a> <span
class="macro">assert_eq!</span>(batch.num_rows(), <span
class="number">3</span>);
+<a href=#783 id=783 data-nosnippet>783</a> <span
class="macro">assert_eq!</span>(batch.num_columns(), <span
class="number">1</span>); <span class="comment">// Only 'data' column
+<a href=#784 id=784 data-nosnippet>784</a>
+<a href=#785 id=785 data-nosnippet>785</a> // Verify the actual values
are preserved after schema evolution
+<a href=#786 id=786 data-nosnippet>786</a> </span><span class="kw">let
</span>data_col = batch.column(<span
class="number">0</span>).as_string::<i32>();
+<a href=#787 id=787 data-nosnippet>787</a> <span
class="macro">assert_eq!</span>(data_col.value(<span class="number">0</span>),
<span class="string">"a"</span>);
+<a href=#788 id=788 data-nosnippet>788</a> <span
class="macro">assert_eq!</span>(data_col.value(<span class="number">1</span>),
<span class="string">"d"</span>);
+<a href=#789 id=789 data-nosnippet>789</a> <span
class="macro">assert_eq!</span>(data_col.value(<span class="number">2</span>),
<span class="string">"g"</span>);
+<a href=#790 id=790 data-nosnippet>790</a> }
+<a href=#791 id=791
data-nosnippet>791</a>}</code></pre></div></section></main></body></html>
\ No newline at end of file
diff --git a/api/src/iceberg/arrow/delete_file_loader.rs.html
b/api/src/iceberg/arrow/delete_file_loader.rs.html
index ac246a63..074bc2d1 100644
--- a/api/src/iceberg/arrow/delete_file_loader.rs.html
+++ b/api/src/iceberg/arrow/delete_file_loader.rs.html
@@ -72,74 +72,78 @@
<a href=#72 id=72 data-nosnippet>72</a> <span
class="prelude-val">Ok</span>(Box::pin(record_batch_stream) <span class="kw">as
</span>ArrowRecordBatchStream)
<a href=#73 id=73 data-nosnippet>73</a> }
<a href=#74 id=74 data-nosnippet>74</a>
-<a href=#75 id=75 data-nosnippet>75</a> <span class="doccomment">///
Evolves the schema of the RecordBatches from an equality delete file
-<a href=#76 id=76 data-nosnippet>76</a> </span><span
class="kw">pub</span>(<span class="kw">crate</span>) <span class="kw">async fn
</span>evolve_schema(
-<a href=#77 id=77 data-nosnippet>77</a> record_batch_stream:
ArrowRecordBatchStream,
-<a href=#78 id=78 data-nosnippet>78</a> target_schema:
Arc<Schema>,
-<a href=#79 id=79 data-nosnippet>79</a> ) -> <span
class="prelude-ty">Result</span><ArrowRecordBatchStream> {
-<a href=#80 id=80 data-nosnippet>80</a> <span class="kw">let
</span>eq_ids = target_schema
-<a href=#81 id=81 data-nosnippet>81</a> .as_ref()
-<a href=#82 id=82 data-nosnippet>82</a> .field_id_to_name_map()
-<a href=#83 id=83 data-nosnippet>83</a> .keys()
-<a href=#84 id=84 data-nosnippet>84</a> .cloned()
-<a href=#85 id=85 data-nosnippet>85</a> .collect::<Vec<<span
class="kw">_</span>>>();
+<a href=#75 id=75 data-nosnippet>75</a> <span class="doccomment">///
Evolves the schema of the RecordBatches from an equality delete file.
+<a href=#76 id=76 data-nosnippet>76</a> ///
+<a href=#77 id=77 data-nosnippet>77</a> /// Per the [Iceberg
spec](https://iceberg.apache.org/spec/#equality-delete-files),
+<a href=#78 id=78 data-nosnippet>78</a> /// only evolves the specified
`equality_ids` columns, not all table columns.
+<a href=#79 id=79 data-nosnippet>79</a> </span><span
class="kw">pub</span>(<span class="kw">crate</span>) <span class="kw">async fn
</span>evolve_schema(
+<a href=#80 id=80 data-nosnippet>80</a> record_batch_stream:
ArrowRecordBatchStream,
+<a href=#81 id=81 data-nosnippet>81</a> target_schema:
Arc<Schema>,
+<a href=#82 id=82 data-nosnippet>82</a> equality_ids: <span
class="kw-2">&</span>[i32],
+<a href=#83 id=83 data-nosnippet>83</a> ) -> <span
class="prelude-ty">Result</span><ArrowRecordBatchStream> {
+<a href=#84 id=84 data-nosnippet>84</a> <span class="kw">let
</span><span class="kw-2">mut </span>record_batch_transformer =
+<a href=#85 id=85 data-nosnippet>85</a>
RecordBatchTransformer::build(target_schema.clone(), equality_ids);
<a href=#86 id=86 data-nosnippet>86</a>
-<a href=#87 id=87 data-nosnippet>87</a> <span class="kw">let
</span><span class="kw-2">mut </span>record_batch_transformer =
-<a href=#88 id=88 data-nosnippet>88</a>
RecordBatchTransformer::build(target_schema.clone(), <span
class="kw-2">&</span>eq_ids);
-<a href=#89 id=89 data-nosnippet>89</a>
-<a href=#90 id=90 data-nosnippet>90</a> <span class="kw">let
</span>record_batch_stream = record_batch_stream.map(<span class="kw">move
</span>|record_batch| {
-<a href=#91 id=91 data-nosnippet>91</a>
record_batch.and_then(|record_batch| {
-<a href=#92 id=92 data-nosnippet>92</a>
record_batch_transformer.process_record_batch(record_batch)
-<a href=#93 id=93 data-nosnippet>93</a> })
-<a href=#94 id=94 data-nosnippet>94</a> });
-<a href=#95 id=95 data-nosnippet>95</a>
-<a href=#96 id=96 data-nosnippet>96</a> <span
class="prelude-val">Ok</span>(Box::pin(record_batch_stream) <span class="kw">as
</span>ArrowRecordBatchStream)
-<a href=#97 id=97 data-nosnippet>97</a> }
-<a href=#98 id=98 data-nosnippet>98</a>}
-<a href=#99 id=99 data-nosnippet>99</a>
-<a href=#100 id=100 data-nosnippet>100</a><span
class="attr">#[async_trait::async_trait]
-<a href=#101 id=101 data-nosnippet>101</a></span><span class="kw">impl
</span>DeleteFileLoader <span class="kw">for </span>BasicDeleteFileLoader {
-<a href=#102 id=102 data-nosnippet>102</a> <span class="kw">async fn
</span>read_delete_file(
-<a href=#103 id=103 data-nosnippet>103</a> <span
class="kw-2">&</span><span class="self">self</span>,
-<a href=#104 id=104 data-nosnippet>104</a> task: <span
class="kw-2">&</span>FileScanTaskDeleteFile,
-<a href=#105 id=105 data-nosnippet>105</a> schema: SchemaRef,
-<a href=#106 id=106 data-nosnippet>106</a> ) -> <span
class="prelude-ty">Result</span><ArrowRecordBatchStream> {
-<a href=#107 id=107 data-nosnippet>107</a> <span class="kw">let
</span>raw_batch_stream = <span
class="self">self</span>.parquet_to_batch_stream(<span
class="kw-2">&</span>task.file_path).<span class="kw">await</span><span
class="question-mark">?</span>;
-<a href=#108 id=108 data-nosnippet>108</a>
-<a href=#109 id=109 data-nosnippet>109</a> <span
class="self">Self</span>::evolve_schema(raw_batch_stream, schema).<span
class="kw">await
-<a href=#110 id=110 data-nosnippet>110</a> </span>}
-<a href=#111 id=111 data-nosnippet>111</a>}
+<a href=#87 id=87 data-nosnippet>87</a> <span class="kw">let
</span>record_batch_stream = record_batch_stream.map(<span class="kw">move
</span>|record_batch| {
+<a href=#88 id=88 data-nosnippet>88</a>
record_batch.and_then(|record_batch| {
+<a href=#89 id=89 data-nosnippet>89</a>
record_batch_transformer.process_record_batch(record_batch)
+<a href=#90 id=90 data-nosnippet>90</a> })
+<a href=#91 id=91 data-nosnippet>91</a> });
+<a href=#92 id=92 data-nosnippet>92</a>
+<a href=#93 id=93 data-nosnippet>93</a> <span
class="prelude-val">Ok</span>(Box::pin(record_batch_stream) <span class="kw">as
</span>ArrowRecordBatchStream)
+<a href=#94 id=94 data-nosnippet>94</a> }
+<a href=#95 id=95 data-nosnippet>95</a>}
+<a href=#96 id=96 data-nosnippet>96</a>
+<a href=#97 id=97 data-nosnippet>97</a><span
class="attr">#[async_trait::async_trait]
+<a href=#98 id=98 data-nosnippet>98</a></span><span class="kw">impl
</span>DeleteFileLoader <span class="kw">for </span>BasicDeleteFileLoader {
+<a href=#99 id=99 data-nosnippet>99</a> <span class="kw">async fn
</span>read_delete_file(
+<a href=#100 id=100 data-nosnippet>100</a> <span
class="kw-2">&</span><span class="self">self</span>,
+<a href=#101 id=101 data-nosnippet>101</a> task: <span
class="kw-2">&</span>FileScanTaskDeleteFile,
+<a href=#102 id=102 data-nosnippet>102</a> schema: SchemaRef,
+<a href=#103 id=103 data-nosnippet>103</a> ) -> <span
class="prelude-ty">Result</span><ArrowRecordBatchStream> {
+<a href=#104 id=104 data-nosnippet>104</a> <span class="kw">let
</span>raw_batch_stream = <span
class="self">self</span>.parquet_to_batch_stream(<span
class="kw-2">&</span>task.file_path).<span class="kw">await</span><span
class="question-mark">?</span>;
+<a href=#105 id=105 data-nosnippet>105</a>
+<a href=#106 id=106 data-nosnippet>106</a> <span class="comment">// For
equality deletes, only evolve the equality_ids columns.
+<a href=#107 id=107 data-nosnippet>107</a> // For positional deletes
(equality_ids is None), use all field IDs.
+<a href=#108 id=108 data-nosnippet>108</a> </span><span class="kw">let
</span>field_ids = <span class="kw">match </span><span
class="kw-2">&</span>task.equality_ids {
+<a href=#109 id=109 data-nosnippet>109</a> <span
class="prelude-val">Some</span>(ids) => ids.clone(),
+<a href=#110 id=110 data-nosnippet>110</a> <span
class="prelude-val">None </span>=>
schema.field_id_to_name_map().keys().cloned().collect(),
+<a href=#111 id=111 data-nosnippet>111</a> };
<a href=#112 id=112 data-nosnippet>112</a>
-<a href=#113 id=113 data-nosnippet>113</a><span class="attr">#[cfg(test)]
-<a href=#114 id=114 data-nosnippet>114</a></span><span class="kw">mod
</span>tests {
-<a href=#115 id=115 data-nosnippet>115</a> <span class="kw">use
</span>tempfile::TempDir;
+<a href=#113 id=113 data-nosnippet>113</a> <span
class="self">Self</span>::evolve_schema(raw_batch_stream, schema, <span
class="kw-2">&</span>field_ids).<span class="kw">await
+<a href=#114 id=114 data-nosnippet>114</a> </span>}
+<a href=#115 id=115 data-nosnippet>115</a>}
<a href=#116 id=116 data-nosnippet>116</a>
-<a href=#117 id=117 data-nosnippet>117</a> <span class="kw">use
super</span>::<span class="kw-2">*</span>;
-<a href=#118 id=118 data-nosnippet>118</a> <span class="kw">use
</span><span class="kw">crate</span>::arrow::delete_filter::tests::setup;
-<a href=#119 id=119 data-nosnippet>119</a>
-<a href=#120 id=120 data-nosnippet>120</a> <span class="attr">#[tokio::test]
-<a href=#121 id=121 data-nosnippet>121</a> </span><span class="kw">async fn
</span>test_basic_delete_file_loader_read_delete_file() {
-<a href=#122 id=122 data-nosnippet>122</a> <span class="kw">let
</span>tmp_dir = TempDir::new().unwrap();
-<a href=#123 id=123 data-nosnippet>123</a> <span class="kw">let
</span>table_location = tmp_dir.path();
-<a href=#124 id=124 data-nosnippet>124</a> <span class="kw">let
</span>file_io = FileIO::from_path(table_location.as_os_str().to_str().unwrap())
-<a href=#125 id=125 data-nosnippet>125</a> .unwrap()
-<a href=#126 id=126 data-nosnippet>126</a> .build()
-<a href=#127 id=127 data-nosnippet>127</a> .unwrap();
-<a href=#128 id=128 data-nosnippet>128</a>
-<a href=#129 id=129 data-nosnippet>129</a> <span class="kw">let
</span>delete_file_loader = BasicDeleteFileLoader::new(file_io.clone());
-<a href=#130 id=130 data-nosnippet>130</a>
-<a href=#131 id=131 data-nosnippet>131</a> <span class="kw">let
</span>file_scan_tasks = setup(table_location);
+<a href=#117 id=117 data-nosnippet>117</a><span class="attr">#[cfg(test)]
+<a href=#118 id=118 data-nosnippet>118</a></span><span class="kw">mod
</span>tests {
+<a href=#119 id=119 data-nosnippet>119</a> <span class="kw">use
</span>tempfile::TempDir;
+<a href=#120 id=120 data-nosnippet>120</a>
+<a href=#121 id=121 data-nosnippet>121</a> <span class="kw">use
super</span>::<span class="kw-2">*</span>;
+<a href=#122 id=122 data-nosnippet>122</a> <span class="kw">use
</span><span class="kw">crate</span>::arrow::delete_filter::tests::setup;
+<a href=#123 id=123 data-nosnippet>123</a>
+<a href=#124 id=124 data-nosnippet>124</a> <span class="attr">#[tokio::test]
+<a href=#125 id=125 data-nosnippet>125</a> </span><span class="kw">async fn
</span>test_basic_delete_file_loader_read_delete_file() {
+<a href=#126 id=126 data-nosnippet>126</a> <span class="kw">let
</span>tmp_dir = TempDir::new().unwrap();
+<a href=#127 id=127 data-nosnippet>127</a> <span class="kw">let
</span>table_location = tmp_dir.path();
+<a href=#128 id=128 data-nosnippet>128</a> <span class="kw">let
</span>file_io = FileIO::from_path(table_location.as_os_str().to_str().unwrap())
+<a href=#129 id=129 data-nosnippet>129</a> .unwrap()
+<a href=#130 id=130 data-nosnippet>130</a> .build()
+<a href=#131 id=131 data-nosnippet>131</a> .unwrap();
<a href=#132 id=132 data-nosnippet>132</a>
-<a href=#133 id=133 data-nosnippet>133</a> <span class="kw">let
</span>result = delete_file_loader
-<a href=#134 id=134 data-nosnippet>134</a> .read_delete_file(
-<a href=#135 id=135 data-nosnippet>135</a> <span
class="kw-2">&</span>file_scan_tasks[<span
class="number">0</span>].deletes[<span class="number">0</span>],
-<a href=#136 id=136 data-nosnippet>136</a>
file_scan_tasks[<span class="number">0</span>].schema_ref(),
-<a href=#137 id=137 data-nosnippet>137</a> )
-<a href=#138 id=138 data-nosnippet>138</a> .<span class="kw">await
-<a href=#139 id=139 data-nosnippet>139</a> </span>.unwrap();
-<a href=#140 id=140 data-nosnippet>140</a>
-<a href=#141 id=141 data-nosnippet>141</a> <span class="kw">let
</span>result = result.try_collect::<Vec<<span
class="kw">_</span>>>().<span class="kw">await</span>.unwrap();
-<a href=#142 id=142 data-nosnippet>142</a>
-<a href=#143 id=143 data-nosnippet>143</a> <span
class="macro">assert_eq!</span>(result.len(), <span class="number">1</span>);
-<a href=#144 id=144 data-nosnippet>144</a> }
-<a href=#145 id=145
data-nosnippet>145</a>}</code></pre></div></section></main></body></html>
\ No newline at end of file
+<a href=#133 id=133 data-nosnippet>133</a> <span class="kw">let
</span>delete_file_loader = BasicDeleteFileLoader::new(file_io.clone());
+<a href=#134 id=134 data-nosnippet>134</a>
+<a href=#135 id=135 data-nosnippet>135</a> <span class="kw">let
</span>file_scan_tasks = setup(table_location);
+<a href=#136 id=136 data-nosnippet>136</a>
+<a href=#137 id=137 data-nosnippet>137</a> <span class="kw">let
</span>result = delete_file_loader
+<a href=#138 id=138 data-nosnippet>138</a> .read_delete_file(
+<a href=#139 id=139 data-nosnippet>139</a> <span
class="kw-2">&</span>file_scan_tasks[<span
class="number">0</span>].deletes[<span class="number">0</span>],
+<a href=#140 id=140 data-nosnippet>140</a>
file_scan_tasks[<span class="number">0</span>].schema_ref(),
+<a href=#141 id=141 data-nosnippet>141</a> )
+<a href=#142 id=142 data-nosnippet>142</a> .<span class="kw">await
+<a href=#143 id=143 data-nosnippet>143</a> </span>.unwrap();
+<a href=#144 id=144 data-nosnippet>144</a>
+<a href=#145 id=145 data-nosnippet>145</a> <span class="kw">let
</span>result = result.try_collect::<Vec<<span
class="kw">_</span>>>().<span class="kw">await</span>.unwrap();
+<a href=#146 id=146 data-nosnippet>146</a>
+<a href=#147 id=147 data-nosnippet>147</a> <span
class="macro">assert_eq!</span>(result.len(), <span class="number">1</span>);
+<a href=#148 id=148 data-nosnippet>148</a> }
+<a href=#149 id=149
data-nosnippet>149</a>}</code></pre></div></section></main></body></html>
\ No newline at end of file
