v0.65.0

[0.65.0] - 2020-03-23

This release includes changes to the client side database schema which will trigger a reset of your blockchain.db and perform full sync on startup.

In this release the txo_list command has reached its full potential in being able to filter specific transaction outputs based on who made them and who controls them which allows for differentiating between "supports" and "tips" and many other use cases. This release also introduces a new txo_sum command which takes almost all of the same arguments as txo_list but produces a sum of the outputs that txo_list would have returned in full, useful for making reports.

Below are some examples and use cases of how to use txo_list.

Get all of the tips you've received from others on claim "xyz" (assuming "xyz" is your claim):
lbrynet txo list --is_my_output --is_not_my_input --type="support" --claim_id="xyz" --unspent

Sum of all of the tips you've sent to claim "xyz" (assuming "xyz" is not your claim):
lbrynet txo sum --is_my_input --is_not_my_output --type="support" --claim_id="xyz"

Note: --unspent is missing above, this is because the local database (at least in desktop app) doesn't know the status of outputs after you've sent them to someone else, therefore filtering by unspent is meaningless here (anytime you pass --is_not_my_output).

Sum of all of the non-tip supports you've sent to claim "xyz" (assuming "xyz" is not your claim):
lbrynet txo sum --is_my_input --is_my_output --type="support" --claim_id="xyz" --unspent

Anything someone just sent me:
lbrynet txo list --is_not_my_input --unspent

Emulate channel list:
lbrynet txo list --type=channel --unspent

Emulate support list:
lbrynet txo list --type=support --unspent

Another flag added is --no_totals which is useful if you expect a very large result set, want a performant pagination and don't need to know total number of pages. (or, a more clever strategy, is to get total once and cache the size for some period of time and then do subsequent calls with --no_totals as user paginates)

Also, --order_by supports a value of none which is useful in certain situations where you don't care about the order and want maximum performance (or, if you just want to see performance impact actually is of sorting the result).

API

• [new feature] added --include_received_tips to claim_list and txo_list commands (#2872) by eukreign
• [new feature] added new flags to resolve (and a few to claim_search) commands: --include_purchase_receipt, --include_is_my_output, --include_sent_supports, --include_sent_tips and --include_received_tips (#2870) by eukreign
• [new feature] txo_list adds many new ownership filters such as --is_my_input, --is_my_output, etc and some metadata filters such as --channel_id, --reposted_claim_id, etc and a new command txo_sum which takes the same arguments as txo_list and produces sum of outputs (#2862) by eukreign
• [new feature] Adds tracemalloc api for memory troubleshooting (#2854) by shyba

Blobs/files

• [bug] fixed issue with shell parsing on Windows (#2866) by BrannonKing
• [refactor] transcode in stream_update, re-enable transcode on Windows (#2861) by BrannonKing

Database

• [new feature] Use multiple processes for querying the db for api calls (#2823) by jackrobison

Discovery

• [new feature] Trending algorithm with time delay and variable decay rate (#2857) by eggplantbren

Files

• [improvement] ffmpeg transcoding - default to 1080P for any source higher than 1080p (#2875) by tzarebczan

Tests

• [improvement] upgrade lbrycrd to v0.17.4.4 (#2865) by eukreign

Wallet

• [bug] fix to prevent transaction cache from leaking some information about outputs between unrelated wallets (#2871) by eukreign
• [improvement] fetch headers on demand (#2863) by shyba

Wallet server

• [new feature] Add blockchain.transaction.info and blockchain.transaction.get_batch to the wallet server for more efficient transaction fetching (#2851) by jackrobison