Merge pull request #328171 from dotlambda/nextcloud

nextcloud28: 28.0.7 -> 28.0.8, nextcloud29: 29.0.3 -> 29.0.4

pkgs/servers/nextcloud/default.nix

@@ -42,14 +42,14 @@ let
   };
 in {
   nextcloud28 = generic {
-    version = "28.0.7";
-    hash = "sha256-WOh1S4Ip/UTr1ykmQlaRlbi2R4xArBCaoIbIa1yrp9k=";
+    version = "28.0.8";
+    hash = "sha256-VaL3RfzI8BtYFrIzM/HjAU0gQKZnlOEy3dDSGaN75To=";
     packages = nextcloud28Packages;
   };
 
   nextcloud29 = generic {
-    version = "29.0.3";
-    hash = "sha256-pZludkwSCSf4hE2PWyjHNrji8ygLEgvhOivXcxzbf9Q=";
+    version = "29.0.4";
+    hash = "sha256-GcRp4mSzHugEAPg5ZGCFRUZWnojbTBX8CFThkvlgJ+s=";
     packages = nextcloud29Packages;
   };
 

pkgs/servers/nextcloud/packages/28.json

@@ -80,7 +80,7 @@
     "hash": "sha256-JsdhEQMSUpnpBFQyDPAVGjT6mAgR826FD3EBOFhsr4c=",
     "url": "https://github.com/nextcloud-releases/integration_openai/releases/download/v2.0.3/integration_openai-v2.0.3.tar.gz",
     "version": "2.0.3",
-    "description": "⚠️ The smart pickers have been removed from this app\nas they are now included in the [Assistant app](https://apps.nextcloud.com/apps/assistant).\n\nThis app implements:\n\n* Text generation providers: Free prompt, Summarize, Headline and Reformulate (using any available large language model)\n* A Translation provider (using any available language model)\n* A SpeechToText provider (using Whisper)\n* An image generation provider\n\nInstead of connecting to the OpenAI API for these, you can also connect to a self-hosted [LocalAI](https://localai.io) instance\nor to any service that implements an API similar to the OpenAI one, for example: [Plusserver](https://www.plusserver.com/en/ai-platform/) or [MistralAI](https://mistral.ai).\n\n## Ethical AI Rating\n### Rating for Text generation using ChatGPT via OpenAI API: 🔴\n\nNegative:\n* the software for training and inference of this model is proprietary, limiting running it locally or training by yourself\n* the trained model is not freely available, so the model can not be run on-premises\n* the training data is not freely available, limiting the ability of external parties to check and correct for bias or optimise the model's performance and CO2 usage.\n\n\n### Rating for Translation using ChatGPT via OpenAI API: 🔴\n\nNegative:\n* the software for training and inference of this model is proprietary, limiting running it locally or training by yourself\n* the trained model is not freely available, so the model can not be run on-premises\n* the training data is not freely available, limiting the ability of external parties to check and correct for bias or optimise the model's performance and CO2 usage.\n\n### Rating for Image generation using DALL·E via OpenAI API: 🔴\n\nNegative:\n* the software for training and inferencing of this model is proprietary, limiting running it locally or training by yourself\n* the trained model is not freely available, so the model can not be ran on-premises\n* the training data is not freely available, limiting the ability of external parties to check and correct for bias or optimise the model’s performance and CO2 usage.\n\n\n### Rating for Speech-To-Text using Whisper via OpenAI API: 🟡\n\nPositive:\n* the software for training and inferencing of this model is open source\n* The trained model is freely available, and thus can run on-premise\n\nNegative:\n* the training data is not freely available, limiting the ability of external parties to check and correct for bias or optimise the model’s performance and CO2 usage.\n\n### Rating for Text generation via LocalAI: 🟢\n\nPositive:\n* the software for training and inferencing of this model is open source\n* the trained model is freely available, and thus can be ran on-premises\n* the training data is freely available, making it possible to check or correct for bias or optimise the performance and CO2 usage.\n\n\n### Rating for Image generation using Stable Diffusion via LocalAI : 🟡\n\nPositive:\n* the software for training and inferencing of this model is open source\n* the trained model is freely available, and thus can be ran on-premises\n\nNegative:\n* the training data is not freely available, limiting the ability of external parties to check and correct for bias or optimise the model’s performance and CO2 usage.\n\n\n### Rating for Speech-To-Text using Whisper via LocalAI: 🟡\n\nPositive:\n* the software for training and inferencing of this model is open source\n* the trained model is freely available, and thus can be ran on-premises\n\nNegative:\n* the training data is not freely available, limiting the ability of external parties to check and correct for bias or optimise the model’s performance and CO2 usage.\n\nLearn more about the Nextcloud Ethical AI Rating [in our blog](https://nextcloud.com/blog/nextcloud-ethical-ai-rating/).",
+    "description": "⚠️ The smart pickers have been removed from this app\nas they are now included in the [Assistant app](https://apps.nextcloud.com/apps/assistant).\n\nThis app implements:\n\n* Text generation providers: Free prompt, Summarize, Headline, Context write, Chat, and Reformulate (using any available large language model)\n* A Translation provider (using any available language model)\n* A SpeechToText provider (using Whisper)\n* An image generation provider\n\nInstead of connecting to the OpenAI API for these, you can also connect to a self-hosted [LocalAI](https://localai.io) instance\nor to any service that implements an API similar to the OpenAI one, for example: [Plusserver](https://www.plusserver.com/en/ai-platform/) or [MistralAI](https://mistral.ai).\n\n## Ethical AI Rating\n### Rating for Text generation using ChatGPT via the OpenAI API: 🔴\n\nNegative:\n* The software for training and inference of this model is proprietary, limiting running it locally or training by yourself\n* The trained model is not freely available, so the model can not be run on-premises\n* The training data is not freely available, limiting the ability of external parties to check and correct for bias or optimise the model's performance and CO2 usage.\n\n\n### Rating for Translation using ChatGPT via the OpenAI API: 🔴\n\nNegative:\n* The software for training and inference of this model is proprietary, limiting running it locally or training by yourself\n* The trained model is not freely available, so the model can not be run on-premises\n* The training data is not freely available, limiting the ability of external parties to check and correct for bias or optimise the model's performance and CO2 usage.\n\n### Rating for Image generation using DALL·E via the OpenAI API: 🔴\n\nNegative:\n* The software for training and inferencing of this model is proprietary, limiting running it locally or training by yourself\n* The trained model is not freely available, so the model can not be ran on-premises\n* The training data is not freely available, limiting the ability of external parties to check and correct for bias or optimise the model’s performance and CO2 usage.\n\n\n### Rating for Speech-To-Text using Whisper via the OpenAI API: 🟡\n\nPositive:\n* The software for training and inferencing of this model is open source\n* The trained model is freely available, and thus can run on-premise\n\nNegative:\n* The training data is not freely available, limiting the ability of external parties to check and correct for bias or optimise the model’s performance and CO2 usage.\n\n### Rating for Text generation via LocalAI: 🟢\n\nPositive:\n* The software for training and inferencing of this model is open source\n* The trained model is freely available, and thus can be ran on-premises\n* The training data is freely available, making it possible to check or correct for bias or optimise the performance and CO2 usage.\n\n\n### Rating for Image generation using Stable Diffusion via LocalAI : 🟡\n\nPositive:\n* The software for training and inferencing of this model is open source\n* The trained model is freely available, and thus can be ran on-premises\n\nNegative:\n* The training data is not freely available, limiting the ability of external parties to check and correct for bias or optimise the model’s performance and CO2 usage.\n\n\n### Rating for Speech-To-Text using Whisper via LocalAI: 🟡\n\nPositive:\n* The software for training and inferencing of this model is open source\n* The trained model is freely available, and thus can be ran on-premises\n\nNegative:\n* The training data is not freely available, limiting the ability of external parties to check and correct for bias or optimise the model’s performance and CO2 usage.\n\nLearn more about the Nextcloud Ethical AI Rating [in our blog](https://nextcloud.com/blog/nextcloud-ethical-ai-rating/).",
     "homepage": "https://github.com/nextcloud/integration_openai"
   },
   "integration_paperless": {
@@ -91,9 +91,9 @@
     "homepage": ""
   },
   "mail": {
-    "hash": "sha256-RWOnUSJBDwSajFafYYv3NnHI93g0lRPKaRU3/Vf8qy8=",
-    "url": "https://github.com/nextcloud-releases/mail/releases/download/v3.7.2/mail-v3.7.2.tar.gz",
-    "version": "3.7.2",
+    "hash": "sha256-zLKTXqXTKyCRrETjKTJzO84QdYar/4zjUrbvT0aWfx0=",
+    "url": "https://github.com/nextcloud-releases/mail/releases/download/v3.7.5/mail-v3.7.5.tar.gz",
+    "version": "3.7.5",
     "description": "**💌 A mail app for Nextcloud**\n\n- **🚀 Integration with other Nextcloud apps!** Currently Contacts, Calendar & Files – more to come.\n- **📥 Multiple mail accounts!** Personal and company account? No problem, and a nice unified inbox. Connect any IMAP account.\n- **🔒 Send & receive encrypted mails!** Using the great [Mailvelope](https://mailvelope.com) browser extension.\n- **🙈 We’re not reinventing the wheel!** Based on the great [Horde](https://horde.org) libraries.\n- **📬 Want to host your own mail server?** We do not have to reimplement this as you could set up [Mail-in-a-Box](https://mailinabox.email)!\n\n## Ethical AI Rating\n\n### Priority Inbox\n\nPositive:\n* The software for training and inferencing of this model is open source.\n* The model is created and trained on-premises based on the user's own data.\n* The training data is accessible to the user, making it possible to check or correct for bias or optimise the performance and CO2 usage.\n\n### Thread Summaries (opt-in)\n\n**Rating:** 🟢/🟡/🟠/🔴\n\nThe rating depends on the installed text processing backend. See [the rating overview](https://docs.nextcloud.com/server/latest/admin_manual/ai/index.html) for details.\n\nLearn more about the Nextcloud Ethical AI Rating [in our blog](https://nextcloud.com/blog/nextcloud-ethical-ai-rating/).",
     "homepage": "https://github.com/nextcloud/mail#readme"
   },
@@ -231,9 +231,9 @@
     "homepage": "https://github.com/nextcloud/user_oidc"
   },
   "user_saml": {
-    "hash": "sha256-cuM5OdrE32gXqK/KiQei+o4SDB0y6mttVtxB+j+vhbc=",
-    "url": "https://github.com/nextcloud-releases/user_saml/releases/download/v6.1.3/user_saml-v6.1.3.tar.gz",
-    "version": "6.1.3",
+    "hash": "sha256-LLjozmU1f8K4KvdKotdTPh3ATG3h2xib/wDP5HGrBsY=",
+    "url": "https://github.com/nextcloud-releases/user_saml/releases/download/v6.2.0/user_saml-v6.2.0.tar.gz",
+    "version": "6.2.0",
     "description": "Using the SSO & SAML app of your Nextcloud you can make it easily possible to integrate your existing Single-Sign-On solution with Nextcloud. In addition, you can use the Nextcloud LDAP user provider to keep the convenience for users. (e.g. when sharing)\nThe following providers are supported and tested at the moment:\n\n* **SAML 2.0**\n\t* OneLogin\n\t* Shibboleth\n\t* Active Directory Federation Services (ADFS)\n\n* **Authentication via Environment Variable**\n\t* Kerberos (mod_auth_kerb)\n\t* Any other provider that authenticates using the environment variable\n\nWhile theoretically any other authentication provider implementing either one of those standards is compatible, we like to note that they are not part of any internal test matrix.",
     "homepage": "https://github.com/nextcloud/user_saml"
   }