v6.2.6

What's Changed

• Changing the gpt-3.5 model to gpt-4o-mini by @mrgick in #242
• chat(): increase the default timeout from 20 to 30 seconds

New Contributors

• @mrgick made their first contribution in #242

Full Changelog: v6.2.5...v6.2.6

v6.2.5

• bump primp to v0.5.5 - Update requirements
• impersonate: add chrome_127
• add def _expand_proxy_tb_alias(): expand "tb" alias to full url
• CLI: chat() - 1) default model gpt-3.5, 2) remove bg="black" in AI answers

Full Changelog: v6.2.4...v6.2.5

v6.2.4

• rename pyreqwest_impersonate to primp
• impersonate: add edge_127, remove chrome_99
• CLI: chat - save|load tokens count to|from json cache

Full Changelog: v6.2.3...v6.2.4

v6.2.3

• chat(): bugfix token counting, rename _chat_conversation_size to _chat_tokens_count

Full Changelog: v6.2.2...v6.2.3

v6.2.2

• Bump pyreqwest_impersonate to v0.5.3, update requirements
• Improve chat:
  1. handle exceptions, add ConversationLimitException,
  2. add _chat_conversation_size (show in CLI as contextsize)

Full Changelog: v6.2.1...v6.2.2

v6.2.1

• DDGS._impersonates: add "safari_17.5"
• _text_html(), _text_lite(): bugfix IndexError
• DDGS._text_lite(): strip() body

Full Changelog: v6.2.0...v6.2.1

v6.2.0

• Update requirements - bump pyreqwest_impersonate to v0.5.0
• chat: small improvement

Full Changelog: v6.1.12...v6.2.0

v6.1.12

What's Changed

• chat(): bugfix (remove superfluous backslashes) by @deedy5 in #238

Full Changelog: v6.1.11...v6.1.12

v6.1.11

What's Changed

• Chat improving (parse response message without json.loads, add timeout parameter) by @deedy5 in #235
• Use orjson if installed, json_dumps(): add ensure_ascii=False, indent=2
• DDGS.init(): randomize impersonate
• Update tests
• README: update

Full Changelog: v6.1.9...v6.1.11

v6.1.9

• Remove orjson from dependencies (tests show that the embedded json module is faster)

Full Changelog: v6.1.8...v6.1.9