english – Kodono

Mise à jour d’un serveur Kimsufi (OVH) depuis Debian 12 (Bookworm) vers Debian 13 (Trixie)

Aymeric — Sun, 17 May 2026 11:03:10 +0000

Cette mise à jour est similaire à ce qui a pu être fait précédemment.

Je vais essayer d’expliquer brièvement les étapes à suivre.

D’abord s’assurer de la version actuelle de Debian avec la commande suivante :
```
lsb_release -a
```
Il est recommandé d’utiliser le programme /usr/bin/script pour enregistrer une transcription de la session de mise à niveau. Ainsi, quand un problème survient, on a un enregistrement de ce qui s’est passé. Pour démarrer un enregistrement, taper :
```
script -t 2>~/upgrade-debian.time -a ~/upgrade-debian.script
```
On va effectuer une mise à jour des paquets :
```
apt-get update && apt-get upgrade
```

On fait une sauvegarde de plusieurs éléments dans le répertoire /root/svg_special :

mkdir /root/svg_special; cp -R /var/lib/dpkg /root/svg_special/; cp /var/lib/apt/extended_states /root/svg_special/; dpkg --get-selections "*" > /root/svg_special/dpkg_get_selection; cp -R /etc /root/svg_special/etc

Si vous utilisez MariaDB, il peut être bien de faire une sauvegarde complète des bases de données :

/usr/bin/mariadb-dump --all-databases --single-transaction -u root -p | gzip > /root/svg_special/backup-databases.sql.gz

Créez un tar.gz du répertoire de la sauvegarde, puis le télécharger dans un lieu sûr :
```
tar -czf /root/backup-pre-debian13.tar.gz /root/svg_special
```
Ensuite il est conseillé d’utiliser screen pour pouvoir se reconnecter (avec screen -r) en cas de déconnexion :
```
screen
```
À noter que si vous n’arrivez pas à scroller dans une session “screen”, il faut alors utiliser la commande CTRL+A, ESC pour cela. Et pour arrêter screen on fait CTRL+A, k
On peut lancer la commande dpkg --audit pour s’assurer que tout est bon avant la migration. On peut également taper dpkg --get-selections "*" | more et vérifier qu’aucun paquet n’est en on hold

On peut supprimer les résidus de fichiers de configuration avec la commande :

find /etc -name '*.dpkg-*' -o -name '*.ucf-*' -o -name '*.merge-error' -delete

Faire un tour dans /etc/apt/ afin de voir si des paquets ne sont pas marqués “bookworm” ou s’ils proviennent de sources tiers. Par exemple /etc/apt/sources.list.d/php.list qui pourrait indiquer deb [signed-by=/etc/apt/keyrings/sury-php.gpg] https://packages.sury.org/php/ bookworm main – pour ce type de paquet ne provenant pas de Debian, on peut commenter la liste avec # le temps de l’upgrade. Rien ne sera désinstallé, et après l’upgrade il faudra s’occuper de ces sources tierces.
Maintenant il faut remplacer tous les “bookworm” de /etc/apt/sources.list par des “trixie”.
On pourra par exemple utiliser la commande suivante :
```
sed -i 's/bookworm/trixie/g' /etc/apt/sources.list
```
On passe aux choses sérieuses, en commençant par mettre à jour les listes des paquets :
```
apt update
```
On va vérifier qu’on a la place suffisante (un message explicite apparait sinon) :
```
apt -o APT::Get::Trivial-Only=true full-upgrade
```
On va maintenant faire une mise à jour minimale :
```
apt upgrade --without-new-pkgs
```
Et à partir de là le système va vous questionner… en général choisir l’option par défaut si vous ne savez pas quoi répondre
Puis on termine avec
```
apt full-upgrade
```

Cette dernière étape va durer un certain temps. Une fois terminé, vous pouvez redémarrer le serveur pour s’assurer que tout va bien.

Réactiver les sources tierces

On va reprendre un par un les fichiers dans /etc/apt/sources.list.d afin de les décommenter. On remplacera “bookworm” par “trixie”, puis on fait apt update && apt upgrade.

MariaDB

On va vérifier la version de MariaDB qu’on a avec mariadb --version. Debian aura installé la version la plus récente qui est disponible dans ses dépôts, mais on peut vouloir utiliser le fichier /etc/apt/sources.list.d/mariadb.sources afin d’avoir des mises à jour plus rapides. Plus d’explications sont disponibles sur le site de MariaDB.

PHP

Il est bien de vérifier que la version actuelle de PHP est correctement utilisée par Apache et qu’elle correspond à ce qu’on veut. Pour cela on vérifie la version avec:

php -v

Ensuite on regarde les versions de PHP disponibles dans les modules d’Apache :

ls -l /etc/apache2/mods-available/php*

Et on regarde celle activée :

ls -l /etc/apache2/mods-enabled/php*

Si la version souhaitée est manquante dans les modules, alors on l’installe, par exemple pour la 8.4 :

apt-get install php8.4 php8.4-mysql

On s’assure ensuite de bien activer la bonne version, par exemple en passant de la v7.4 à v8.4 :

a2dismod php7.4
a2enmod php8.4

Et on redémarre Apache :

systemctl restart apache2

Pour finir

Une fois les erreurs corrigées, on va nettoyer tous les paquets avec :

apt-get autoremove

Make a field ReadOnly in SharePoint List

Aymeric — Tue, 28 Apr 2026 09:26:32 +0000

We can make a field readonly or editable in SharePoint using REST API. I usually use it when I want to update a bunch of SharePoint List Items without changing their Modified date and Editor details. Below is an example to make “Editor” readonly again (you’ll first need to get the Digest Token):

fetch(`${siteUrl}/_api/web/lists/getByTitle('${listName}')/fields/getbyinternalnameortitle('Editor')`, {
    method: "POST",
    headers: {
      "Accept": "application/json;odata=verbose",
      "Content-Type": "application/json;odata=verbose",
      "X-HTTP-Method": "MERGE",
      "IF-MATCH": "*",
      "X-RequestDigest": requestDigest
    },
    body: JSON.stringify({
      "__metadata": { "type": "SP.FieldUser" },
      "ReadOnlyField": true
    })
  });

This method works even if the list has more than 5,000 items.

To verify if a field is readonly or not, we can simply use the endpoint `${siteUrl}/_api/web/lists/getByTitle('${listName}')/fields/getbyinternalnameortitle('Editor')?$select=ReadOnlyField`

Issues with the Chrome extension “Microsoft Single Sign On”

Aymeric — Mon, 13 Apr 2026 06:56:11 +0000

When you use Chrome on a corporate network, you need an extension called “Microsoft Single Sign On” to be correctly authenticated with the Microsoft 365 websites (like SharePoint, …).

Recently the extension failed multitimes and I was not able to connect to my Microsoft work account anymore using Chrome. I had to clear cookies, restart the computer, etc.

By searching on reddit, I found it’s now possible to get ride of the extension because Chrome can natively handle this type of connection. To proceed, you need to edit the Windows Registry (Computer\HKEY_LOCAL_MACHINE\SOFTWARE\Policies\Google\Chrome) to add the REG_DWORD (DWORD (32-bit) Value) called “CloudAPAuthEnabled” and put the value 1, as explained in the Chrome documentation.

It’s been a couple of weeks I’m using it, and no more issues!

Power Automate Flow: copy folder when one of its files is updated

Aymeric — Mon, 09 Feb 2026 15:10:00 +0000

I had the scenario where when files are updated in a specific folder of a SharePoint Documents Library, then the folder should be copied to another document library.

Because we don’t want the copy process to trigger for ever changes (like if we bulk upload files, or if someone is doing changes in live in a file), so here is my solution:

Create a SharePoint List called “File Processing” with two fields:
- A lookup field called “File” that is based on our source Document Library – make sure to select the option “Enforce unique values”;
- A Date and Time field called “LastModified”.
Create a Power Automate Flow that is manually triggered – we’ll call it with “Run a Child”: this one is only used to do the copy of the folder from source to destination
Create a Power Automate Flow that triggers when a file is created or modified (properties only):
1. When we confirm the file updates happened in the correct folder, then we add an entry in the “File Processing” list with the file ID and the file Modified date;
2. Terminate the flow.
Create a Power Automate Flow that triggers when an item is created or modified in the “File Process” list:
1. Add a “Delay” of 5 minutes;
2. Add the action “Get file properties” from our source document library using the file ID from the item;
3. Check with a condition if the field “LastModified” is greater or equal to the Modified date of the file:
  - If yes, then use the action “Get items” on the “File Processing” list to see if we have any other items in the list: if no other items, run the child process to make the copy – and, finally, delete the current item in “File Processing” whatever there is other items or not;
  - If no, then update the field “LastModified” of the current item using the Modified date of the file.

This solution permits to only trigger the copy process when there is no more files update in the folder.

Note: if we nee to monitor different folders in the source library, we can add a lookup column in “File Process” that contains the folder ID, and when checking if there is no other items for the folder, we just add a condition based on the folder ID.

SharePoint REST API to copy a file from one site to another one

Aymeric — Tue, 03 Feb 2026 16:31:54 +0000

There is a good explanation in this Gist, and make sure to read my comment also!

Set People Picker field in a SharePoint List Item using Microsoft Graph

Aymeric — Mon, 02 Feb 2026 11:50:50 +0000

When we want to use MS Graph to add an item into a SharePoint List that contains a People Picker, then here what needs to be done:

Retrieve the Site ID for the root site collection, using https://graph.microsoft.com/v1.0/sites/[CORP].sharepoint.com:/sites/MySiteCollection
Retrieve the Site ID for the target end website, using https://graph.microsoft.com/v1.0/sites/[CORP].sharepoint.com:/sites/MySiteCollection/MyParentSiteWeb/MyTargetWebsite/
Use the user’s email address to retrieve the SharePoint User ID of that person from the User Information List, using https://graph.microsoft.com/v1.0/sites/[ROOT_SITE_ID]/lists/User%20Information%20List/items?expand=fields&filter=fields/EMail%20eq%20'[ENCODED_EMAIL]'&top=1" – the email must be encoded to be passed in an URL
If the user doesn’t exist in this SharePoint yet, then you have to use the REST API call to https://[CORP].sharepoint.com/sites/MySiteCollection/MyParentSiteWeb/MyTargetWebsite/_api/web/ensureUser with the payload {"logonName":"user_email@corp.com"} – you should get the User ID this time
Then we need to get the List ID of where we want to add the item, using https://graph.microsoft.com/v1.0/sites/[TARGET_SIDE_ID]/lists/[LIST_NAME_ENCODED]
Finally, we can add the item using https://graph.microsoft.com/v1.0/sites/[TARGET_SIDE_ID]/lists/[TARGET_LIST_ID]/items with a specific value depending if it’s a multi-selection people picker field or not – let’s say the internal name for my field is “Requestor_x0020_Name”, then I have to append “LookupId” to this name:
- for a single selection field, the payload contains:
```
{"Requestor_x0020_NameLookupId": "[USER_ID_FOUND_WITH_GRAPH]"}
```
- for a multi-selection field we have to pass the odata type, and put the User ID into an array, then the payload contains:
```
{
"Requestor_x0020_NamesLookupId@odata.type":"Collection(Edm.String)",
"Requestor_x0020_NamesLookupId": [ "[USER_ID_FOUND_WITH_GRAPH]" ]
}
```

How to pass a parameter from the URL to a SharePoint Form (2025)

Aymeric — Sun, 28 Sep 2025 16:33:07 +0000

For a few months, there is a new way to display a new form for a SharePoint List, using Microsoft Forms:

This kind of form can be useful for intake requests or surveys. But how to pass a parameter in the URL to apply it to a field in the form?

Here is the solution I deployed:

Create a Power Automate Flow that will replace the default value of the field with a defined value
Create a Power App to intercept the parameter from the URL, to trigger the above flow, and to redirect the user to the form
The user will see the defined value in the field because we use the default value
A few seconds later, the flow will change back the default value to blank

It works well. The only downside is that we cannot have more than 1 person clicking on the link at the same time (and within a range of 15 seconds).

1) The Power Automate Flow

The flow is called by a Power App with a string parameter.

Then we do an HTTP call to SharePoint with the below details:

Method: PATCH
Uri: _api/web/lists/GetByTitle(‘SHAREPOINT LIST NAME’)/Fields/GetByInternalNameOrTitle(‘FIELD_ID’)
Headers: “Accept: application/json;odata=nometadata”, “Content-Type: application/json;odata=nometadata”, and “IF-MATCH: *”
Body: we use a JSON {"DefaultValue": "VALUE PROVIDED BY APP"}

By default value, I mean this information (legacy SharePoint interface):

Once the default value of the field has been updated, we respond back to the Power App.

We add a 15-second delay.

And finally, we do another HTTP call to SharePoint, with the same info, except for the body that becomes {"DefaultValue":""} to reset the default value.

2) The Power App

Instead of giving the direct link to the form to the user, we’ll give them a link to the Power App. The URL must contains the param value (e.g. “&Title=Code_123456”).

The App can have a short text (e.g. “click the below button to get to the survey”) with a button “Go to the form”.

When the user clicks on the button, it will trigger the flow we created in step 1, and then it will redirect the user to the form.

To do so, for OnSelect on the button:

// we can change the aspect of the button to indicate something is happening
Set(buttonLabel, "Loading survey…");
Set(buttonMode, DisplayMode.Disabled);

// decode the "Title" param that is passed in the URL
Set(
  titleParam,
  Substitute(
    Substitute(
        Substitute(
        Substitute(
            Substitute(Param("Title"), "%3A", ":"),
            "%2F", "/"
        ),
        "%3F", "?"
        ),
        "%3D", "="
    ),
    "%26", "&"
    )
);

// call the Power Automate Flow
'PowerApps–SetURLParamasDefaultValue'.Run(titleParam);

// then redirect to the form
Launch(urlToForm, {}, LaunchTarget.Replace);

3) Let’s the magic happen

When the user clicks on the button:

the flow runs,
the flow changes the default value using the provided param,
the flow says to the app to proceed,
the app redirects the user to the form,
the form shows up with the field that already contains the provided value
15 seconds later (the time for the form to load), the flow resets the default value

AndroidManifest.xml from ADB

Aymeric — Sun, 27 Apr 2025 14:33:12 +0000

It’s possible to find the AndroidManifest.xml file for an app using its APK (e.g. from APKMirror) and then with the help of a tool like jadx, but we can also connect to the Android device using adb connect, and then we use adb shell dumpsys package PACKAGE_NAME (e.g. “com.netflix.ninja” is the package name for Netflix).

This command will return many information similar to what the AndroidManifest.xml can provide.

Power FX: Dynamically access an object property with a variable

Aymeric — Thu, 03 Apr 2025 16:31:58 +0000

I’m kind of new with Power Apps, and it’s been very difficult to find how to access an object property using a variable.

In a nutshell, here are to access it:

With(
  { tempData: ParseJSON(JSON(ThisItem, JSONFormat.IgnoreBinaryData & JSONFormat.IgnoreUnsupportedTypes)) },
  Concat(
    ForAll(
      ColumnArray As ColumnName,
      ColumnName.Value & "=>" & Column( tempData, ColumnName.Value ) & "
"
    ),
    Value
  )
)

I’m transforming my record ThisItem to an UntypedObject to make it work with Column(), using JSON and by removing the problematic properties (thanks to JSONFormat.IgnoreBinaryData & JSONFormat.IgnoreUnsupportedTypes).

Then I iterate my array (a.k.a ColumnArray) that contains the property I need, and I use Column( tempData, ColumnName.Value ) to extract the column from the record!

My use case: I defined a collection with a list of columns, then I have a Gallery in which I have an HTMLText component that must display the columns from ThisItem based on the collection I defined before.

So, in App screen, I defined my collection like that:

ClearCollect(ColumnArray,["ID", "Title", "Description", "Target_x0020_Start_x0020_Date", "AssignedTo"]);
ClearCollect(UserColumns, ["AssignedTo"]);

Then in my Gallery, I have an HTMLText object with the below content:

"" &
With(
  { tempData: ParseJSON(JSON(ThisItem, JSONFormat.IgnoreBinaryData & JSONFormat.IgnoreUnsupportedTypes)) },
  Concat(
    ForAll(
      ColumnArray As ColumnName,
      "" &
  If(
    ColumnName.Value in UserColumns,
    Column( tempData, ColumnName.Value ).DisplayName,
    Column( tempData, ColumnName.Value )
  )
  & ""
    ),
    Value
  )
)
& ""

This way I can make my display dynamic, based on a list of columns (that can be populated using a Power Automate Flow). I’ll have to identify which columns are User type, or similar ones, in order to access to the next property (like .DisplayName).

Remove lazy loading for a custom SPFx webpart

Aymeric — Sun, 15 Dec 2024 22:09:45 +0000

SharePoint Online is doing lazy loading when we hit a page with webparts. If you created a SPFx webpart, and if it’s not visible right after the page load, then it will only be loaded once the user scrolls to it…

To remove this behavior, and load the webpart as soon as possible, you must declare your webpart as a dynamic data source.

Here is the minimal code example (in pure JS, not in TS):

import { BaseClientSideWebPart } from '@microsoft/sp-webpart-base';
import { DisplayMode } from '@microsoft/sp-core-library';

export default class HtmlViewerWebPart extends BaseClientSideWebPart {
  constructor() {
    super();
    // the below is used by the dynamic data source to avoid lazy loading
    this._sourceId = 'my_webpart';
  }

  onInit() {
    // to avoid lazy loading of the webpart, we register it as a dynamic data source
    // we only need to do it during the Edit of the page
    if (this.isPageInEditMode() && this.context.dynamicDataSourceManager) {
      this.context.dynamicDataSourceManager.initializeSource(this);
      console.log('Dynamic Data Source initialized to avoid lazy loading.');
    }
    
    return Promise.resolve();
  }

  isPageInEditMode() {
    return this.displayMode === DisplayMode.Edit;
  }

  // we need to declare this method because it's used by the data source to avoid lazy loading
  getPropertyDefinitions() {
    return []
  }

  // we need to declare this method because it's used by the data source to avoid lazy loading
  getPropertyValue(propertyId) {
    return propertyId;
  }
}

Then, edit the page, which will trigger the dynamic data source registration, and publish it.

Determine an element `height` and `width` in CSS only and reuse it within another CSS rule

Aymeric — Fri, 13 Dec 2024 15:44:36 +0000

I found this technic on https://frontendmasters.com/blog/how-to-get-the-width-height-of-any-element-in-only-css/ – it only works on modern browsers (mainly Chrome and Edge)

Let’s define the properties:

@property --_x {
  syntax: "";
  inherits: true;
  initial-value: 0; 
}
@property --_y {
  syntax: "";
  inherits: true;
  initial-value: 0; 
}
@property --w {
  syntax: "";
  inherits: true;
  initial-value: 0; 
}
@property --h {
  syntax: "";
  inherits: true;
  initial-value: 0; 
}

Then, let’s find an element that is a common parent for both the element where we’ll calculate the height/width, and the element that will use this height/width, and we apply some CSS:

.parent {
  timeline-scope: --cx,--cy;
  --w:calc(1/(1 - var(--_x)));
  --h:calc(1/(1 - var(--_y)));
  animation: x linear,y linear;
  animation-timeline: --cx,--cy !important;
  animation-range: entry 100% exit 100%;
}

We finish up with some additional CSS:

.get-dimension {
  overflow: hidden;
  position: relative;
}
.get-dimension:before {
  content:"";
  position: absolute;
  left: 0;
  top: 0;
  width: 1px;
  aspect-ratio: 1;
  view-timeline: --cx inline,--cy block;
}
@keyframes x {to{--_x:1}}
@keyframes y {to{--_y:1}}

And finally, we can use --w and --h in the other element:

.apply-dimension {
  height: calc(var(--h)*1px);
  width: calc(var(--w)*1px);
}

Here is a demo:

How to remove an account / leave an organization in MS Teams?

Aymeric — Fri, 11 Oct 2024 15:44:15 +0000

If in the top right corner of MS Teams, when you click on your profile picture, you see other organizations/accounts and you want to delete them:

Go to the MS Teams Settings by clicking on the three dots next to your profile picture
Go to Accounts and orgs
Turn off or leave the org/account that is problematic

How to run Android TV 9 on Windows

Aymeric — Mon, 02 Sep 2024 09:34:00 +0000

I used to use Android Studio with the various emulators to test my Android apps, however for Android TV 9 it doesn’t work properly because it doesn’t have Google Play Store on it.

To have a good version of Android TV 9:

Downlad a VirtualBox image of it – I used the one shared in this video
Install VirtualBox if you don’t have it yet
Load the downloaded image with VirtualBox
Configure it by going to the network section and select “Bridged Adapter” instead of “NAT”
Once Android TV is started, you’ll have to connect to your Google account, but using the keyboard might be challenging – you can right click on the USB cable icon that appears at the bottom right of the VirtualBox window, and select your keyboard (be aware it’s a QWERTY map that is used)

After that, go to the Android TV settings, in the “About” section, and click several times on the Build Version to enable the Developer Mode. From the developer menu, you can enable the “Debug USB”.

Once everything is correctly configured, the Android TV should be visible on your local network, meaning you can use the ADB command to test your app.

Manual Installation of ExplorerPatcher

Aymeric — Thu, 21 Mar 2024 08:59:15 +0000

ExplorerPatcher permits to change the behavior of the Windows taskbar. If for some reasons the automatic install doesn’t work, you can proceed manually:

Install it on a computer where the automatic install works
On the target computer, download ep_setup.exe from the official website
Open a Terminal and extract the files from the setup using the command ep_setup.exe /extract
Several files have been extracted from the setup
Move all the files to C:\Program Files\ExplorerPatcher
From the computer in step 1, transfer the file C:\Windows\dxgi.dll to your other computer
Use a CMD as an Administrator, then kill Windows Explorer, paste the dxgi.dll into C:\Windows, and relaunch Windows Explorer:
```
taskkill /f /im explorer.exe && copy dxgi.dll C:\Windows\ && explorer.exe
```

To manually uninstall it:

Open a CMD as an Administrator, then type: taskkill /f /im explorer.exe && del C:\Windows\dxgi.dll && rmdir /s /q "C:\Program Files\ExplorerPatcher" && explorer.exe
Open regedit as an Administrator, and delete HKEY_CURRENT_USER\Software\ExplorerPatcher, and search for any other references of ExplorerPatcher to delete
Delete the forlder %APPDATA%\ExplorerPatcher
Restart the computer

Power Automate: how to verify if a property belongs to an object (apply to SharePoint Date too)

Aymeric — Tue, 16 Jan 2024 17:36:30 +0000

We can use this kind of formula (notice the questionmark):
if(empty(variables('params')?[variables('fieldName')]), 'fieldName is not part of the object params', variables('params')?[variables('fieldName')])

Then it either returns “fieldName is not part of the object params”, or the value.

We can use it to check if a date field is empty in a SharePoint List, because when getting the data from the SP List, when the date field is empty, it’s not in the result returned – the below code will return true if the date field is empty:
empty(item()?['dateFieldNameId'])

For a Currency/Number field you need to use string() as well:
empty(string(item()?['floatFieldNameId']))

Power Automate returns an error about “InvokerConnectionOverrideFailed” and “header.X-MS-APIM-Tokens”

Aymeric — Tue, 16 Jan 2024 14:42:28 +0000

While calling a “Run a Child Flow” from a Power Automate Flow, you could get an error about “InvokerConnectionOverrideFailed” and “header.X-MS-APIM-Tokens”.

After investigating, to resolve this issue you need to open the “details” view of your child flow, and click on the “Edit” button from the “Run only users” card:

Then in the Connections Used section, make sure to select “Use this connection (your_username@domain.com)” instead of the default “Provided by run-only user”. It will permit to run this child flow with your rights.

Search and restore an item from a SharePoint Online Recycle Bin

Aymeric — Tue, 07 Nov 2023 11:09:37 +0000

It might be difficult to search for an item in a SharePoint recycle bin. I was using the end point _api/site/RecycleBin as a Site Collection Administrator, but in some cases it returns an error “The attempted operation is prohibited because it exceeds the list view threshold.”.

The solution is to use another end point _api/site/getrecyclebinitems (see the documentation for the various parameters):

fetch("https://tenant.sharepoint.com/sites/YourSite/_api/site/getrecyclebinitems?rowLimit=%2770000%27&itemState=0&orderby=3&isAscending=false", {
  "headers": {
    "accept": "application/json",
    "content-type": "application/json",
    "x-requestdigest": document.querySelector("#__REQUESTDIGEST").value
  },
  "method": "GET",
})
.then(res => res.json())
.then(json => {
  console.log(json.value);
  // it will show an array with an object that contains several interesting properties:
  // AuthorEmail, AuthorName, DeletedByEmail, DeletedDate, DirName, Id, LeafName, Title, …
})

You can then filter the result to get what you need, for example if you’re looking for an item from the list “Projects” that has the ID 1981:

json.value.filter(item => item.DirName.includes("Lists/Projects") && item.LeafName === "1981_.000");

Then, to restore an item, we need the Id from the previous result. The famous endpoint to restore an item is _api/site/RecycleBin('Id')/restore, however it could also return the error “The attempted operation is prohibited because it exceeds the list view threshold”. In that case, we can use this other endpoint _api/site/RecycleBin/RestoreByIds:

fetch("https://tenant.sharepoint.com/sites/YourSite/_api/site/RecycleBin/RestoreByIds", {
  "headers": {
    "accept": "application/json",
    "content-type": "application/json",
    "x-requestdigest": document.querySelector("#__REQUESTDIGEST").value
  },
  "method": "POST",
  "body":JSON.stringify({
    "ids": [
      "4f855ee7-472b-414a-a482-4317a114c1a2" // Id to restore
    ],
    "bRenameExistingItems": true
  })
})

Connect to SharePoint Online using an app clientId and clientSecret

Aymeric — Tue, 21 Mar 2023 09:56:30 +0000

Get `clientId` and `clientSecret`

(source)

You’ll need credentials:

`clientId` – required string, client id obtained when registering the addin
`clientSecret` – required string, client secret obtained when registering the addin
`realm` – your SharePoint Online tenant id. The easiest way to find tenant is to open SharePoint Online site collection, click Site Settings → Site App Permissions. Under this page you wll see at least one app “Microsoft.SharePoint”. The tenant id (realm) is highlighted in the image below:

Example of the expected result:

{
  clientId: '28bq7e56-8c3a-487d-hbfb-ef1a74539cbe',
  clientSecret: 's6LZ4VvoeKOS+MyAhklcavsyJBF4XhWo06OgY6czYJ0=',
  realm: '85e5f09b-4c17-4d80-afea-260bb171c456'
}

To get the credentials, you need to register a new addin inside SharePoint Online, by fellowing these steps:

Open SharePoint Online app registration page, e.g. https://contoso.sharepoint.com/sites/dev/_layouts/15/appregnew.aspx
Click on “Generate” for Client id and Client Secret, fill in Title, App Domain, Redirect URI (you can type in any values you want)
Click on “Create” and save generated Client Id and Client Secret
[IF YOU HAVE TENANT RIGHTS] Now you need to apply permissions to the newly registered app. If you want to register the app once and use it for any site collection, it’s better to apply tenant scope permissions, so you can use the credentials everywhere inside your SharePoint tenant. To apply tenant scoped permissions, open AppInv.aspx page under SharePoint adminstration web site, e.g. https://[YOUR_ORGANIZATION]-admin.sharepoint.com/_layouts/15/appinv.aspx, copy paste Client Id from step n°3 into App Id field and click “Lookup”.
[IF YOU HAVE TENANT RIGHTS] You will see your registered app, paste in the following XML into the “Permission Request XML” field and click “Create”:
```
    
      
    
```
[IF YOU ARE NOT A TENANT] If you only want to give permissions on 1 site collection, you can register the app on a regular site collection by using url https://contoso.sharepoint.com/sites/dev/_layouts/15/appinv.aspx. In this case you are not able to use tenant scoped permissions and can only apply site collection permissions:
```
    
      
    
```
You will see addin “Trust” confirmation, click on “Trust It”:

if trust-it button is not enabled and you get a red label saying tenant admin needs to trust the app, go back and try again in a few minutes.
Now you can use client id and client secret to send authenticated http requests.

To know more about the XML permissions, you can check the Microsoft documentation.

Get Access Token

(you can find a C# code as an example)

You need to do a POST request to https://accounts.accesscontrol.windows.net/[YOUR_TENANT_REALM]/tokens/OAuth/2 with a “Content-Type” header that has the value “application/x-www-form-urlencoded”, and the body parameters that must be:

"grant_type":"client_credentials"
"client_id":"[YOUR_CLIENT_ID]@[YOUR_TENANT_REALM]"
"client_secret":"[YOUR_CLIENT_SECRET]"
"resource":"00000003-0000-0ff1-ce00-000000000000/dell.sharepoint.com@[YOUR_TENANT_REALM]"

See below an example in PHP:

$curl = curl_init();
curl_setopt($curl, CURLOPT_URL, "https://accounts.accesscontrol.windows.net/[YOUR_TENANT_REALM]/tokens/OAuth/2");
curl_setopt($curl, CURLOPT_HTTPHEADER, [ "Content-Type: application/x-www-form-urlencoded" ]);
curl_setopt($curl, CURLOPT_POSTFIELDS, http_build_query([
  "grant_type" => "client_credentials",
  "client_id" => "[YOUR_CLIENT_ID]@[YOUR_TENANT_REALM]",
  "client_secret" => "[YOUR_CLIENT_SECRET]",
  "resource" => "00000003-0000-0ff1-ce00-000000000000/dell.sharepoint.com@[YOUR_TENANT_REALM]"
]));
curl_setopt($curl, CURLOPT_RETURNTRANSFER, true);
$response = json_decode(curl_exec($curl));
curl_close($curl);

echo $response->access_token;

The response should contain an access token. Example:

{
  "token_type":"Bearer",
  "expires_in":"86399",
  "not_before":"1679393911",
  "expires_on":"1679480611",
  "resource":"00000003-0000-0ff1-ce00-000000000000/dell.sharepoint.com@[YOUR_TENANT_REALM]",
  "access_token":"eyJ0eXAiOiJKV1QiLCJhbGciOiJSU[...]SxXA5Lqbk1OcOVdwQ"
}

Finally, you can do your REST API request to SharePoint Online with passing the header “Authorization” that has the value “Bearer [YOUR_ACCESS_TOKEN]”.

Deploy a PCF NodeJS app as a scheduled task

Aymeric — Tue, 27 Dec 2022 15:23:33 +0000

I have a NodeJS app that runs as a process and that executes a task every 15 minutes using node-schedule.

We first need a manifest.yml file that contains:

---
applications:
- name: APP-NAME
  buildpack: nodejs_buildpack
  no-route: true
  health-check-type: process
  env:
    OPTIMIZE_MEMORY: true

The no-route parameter is true so that we don’t get a route assigned, and the health-check-type is set to process so that the orchestrator monitors process availability and doesn’t try to ping a non-existent web endpoint. And OPTIMIZE_MEMORY in “env” section is based on the Pivotal recommendations.

If you need to use a local package in your app, you’ll have to pack it up first. To do it, go to your local module folder, and type npm pack. It will create a .tgz file that you’ll have to store in a local_modules folder for your app. Next, use npm install .\local_modules\package-1.2.3.tgz.

You can now deploy your app with pcf push APP-NAME and you can read the logs with cf logs APP-NAME --recent.

Power Automate: execute a SQL Query via On-Promise Gateway

Aymeric — Fri, 09 Dec 2022 13:28:26 +0000

In Power Automate, when you want to connect to a SQL Server and if you have a On-Promise Gateway, then you cannot use the command “Execute a SQL Query” because it will say it’s not currently supported.

There is a workaround with “Transform data using Power Query” (ATTENTION: you cannot load it from a flow from a Solution… you’ll have to go to your Flows and edit the flow from there):

Let’s say we have 3 tables: ITEM_CATALOG, CATALOG and CURRENCY. We want to join them and filter them based on a variable found previously in our flow.

First, we can define our where. Here I have several values that I want to test using a IN:

I create a string with my different values separated by a coma.

Next, we can open the Power Query editor:

In the interface, we choose the 3 tables we need to merge and we add a parameter called “where”:

We rename it to “where” and leave the default settings:

Then we use the “Advance Editor”:

And we wrote the below:

let
  where = Text.Split( "@{variables('where')}" , ",")
in
  where

It means we want to split the variable “where” coming from the flow, based on the coma separator:

We can now merge the tables and add a filter:

And when the step to filter is here, we select “in” and our query:

Last step is to “Enable Load” to make sure this is what the operation will return to our flow:

You can run it to test and see if it works.

Then, to get the output from it, we’ll use a “Parse JSON”… The schema is probably something like:

{
    "type": "object",
    "properties": {
        "resultType": {
            "type": "string"
        },
        "value": {
            "type": "array",
            "items": {
                "type": "object",
                "properties": {
                    "COLUMN_A": {
                        "type": "string"
                    },
                    "COLUMN_B": {
                        "type": "integer"
                    },
                    "COLUMN_C": {
                        "type": "string"
                    }
                },
                "required": [
                    "COLUMN_A",
                    "COLUMN_B",
                    "COLUMN_C"
                ]
            }
        }
    }
}

You may need to make several tries in order to find the correct schema. You can also use the “Generate from sample” by pasting the data from the previous step:

We use “value” in the loop:

And then we can access our columns:

Pass an URL parameter to a SharePoint Online form’s field

Aymeric — Wed, 30 Nov 2022 08:43:25 +0000

The only way to pass a URL parameter to a SharePoint Online (modern design) form’s field is to use PowerApps (at least, if you cannot add any JS on your website!).

Important warning: when you use PowerApps to manage your form, all edits to the list settings won’t reflect to the PowerApps form. For example, if you add a field, it won’t show up, and you’ll have to manually update the PowerApps to add it (see at the bottom of this article).

From the list view, go to Integrate then PowerApps and Customize forms:

Once PowerApps has open the form, you need to do several things.

1. Load the ID

We first need to make sure the form will load the required item when we pass the ID URL parameter:

From the SharepointForm Advanced Settings, we change the DefaultMode to check if we have the ID parameter, and if we don’t have it, then it should be a New Form, otherwise an Edit Form:

If(IsBlank(Param("ID")), FormMode.New, FormMode.Edit)

From the SharepointForm Advanced Settings, we change the Item section to check if we have the ID parameter, and if we have it, then we do a lookup in our list to find the data:

If(IsBlank(Param("ID")), SharePointIntegration.Selected, LookUp('NAME OF THE LIST', ID = Value(Param("ID"))))

Add a SUBMIT button

With PowerApps, there is no button to save the changes! We’ll add a button in the form:

In the button’s properties, we change the onSelect to be:

SubmitForm(SharePointForm1)

Be aware that the page will stay with the form after clicking on the button. You could want to close using the Exit() function, but the user will be redirected on office.com … I’d recommend to use Launch() by redirecting your users to a page:

SubmitForm(SharePointForm1); Launch("https://tenant.sharepoint.com/sites/MySite/");

Set field’s value based on URL parameter

We can finally set the field’s value based on the parameter in the URL. Select the INPUT zone of the field, and in the Default section we use the below formula:

If(IsBlank(Param("Title")), ThisItem.Title, Param("Title"))

Here my field is called “Title” so I decided to use a parameter called “Title” as well.

Link to the form

We cannot use the NewForm.aspx or EditForm.aspx to access this form, but we need a special link.

Go to your list settings:

Then go to the form settings (it’s from there that you can decide to keep PowerApps or use the original Sharepoint Forms), and click on See versions and usage:

You’ll get the App Id from this page:

Next, you’ll use the App Id to forge the URL: https://apps.powerapps.com/play/providers/Microsoft.PowerApps/apps/APP_ID
With our example, the URL will be: https://apps.powerapps.com/play/providers/Microsoft.PowerApps/apps/c6f23ac1-dcbd-4daf-925e-2701ab241ca0

You can now pass the URL parameter: https://apps.powerapps.com/play/providers/Microsoft.PowerApps/apps/APP_ID?Title=Hello%20World
And an ID to retrieve an existing item: https://apps.powerapps.com/play/providers/Microsoft.PowerApps/apps/APP_ID?Title=Hello%20World&ID=2

How to use it with a LookUp column?

If you want to auto-select a LookUp field using an URL parameter, you need to do a few things…

First, we need to add the related table. To do so, click on Data in the left navigation bar and search for SharePoint:

Search for the table and add it.

Second (optional) step: click on the Lookup field in the form and change the Items to show a list of options – if no “Lookup” ID in the URL, then we use the default list of options:

The below formula permits to retrieve the “ID” and “Title” from the distant list, base on the “Lookup” parameter, and to rename the result as {Id:"ID", Value:"Title"}:

If(IsBlank(Param("Lookup")), Choices([@'CURRENT LIST NAME'].COLUMN_NAME), RenameColumns(ShowColumns(Filter('DISTANT LIST NAME', ID = Value(Param("Lookup"))), "ID", "Title"), "ID", "Id", "Title", "Value"))

Third, click on the Lookup field in the form and change the DefaultSelectedItems to select the item from the list of options:

The below formula returns an empty selection with {Id:"", Value:""} when no URL param, otherwise it returns the first record for our lookup:

If(IsBlank(Param("Lookup")), {Id:"", Value:""}, First(RenameColumns(ShowColumns(Filter('DISTANT LIST NAME', ID = Value(Param("Lookup"))), "ID", "Title"), "ID", "Id", "Title", "Value")))

And finally, we can pass Lookup=ID in the URL to select the related item in the other list

How to deal with new fields?

If you add a new field to your list’s settings, you’ll have to edit the form in PowerApps, and then edit the fields and add the new one:

(I used this article as a starting point)

Transfer an Alexa AWS Lambda function from the online editor to the ASK CLI

Aymeric — Tue, 20 Sep 2022 17:59:44 +0000

When we follow the guide to build a new smarthome skill, it gives the steps to create a function in the online code editor.

But if you prefer to use the ASK CLI, there is some steps to follow…

I first create a fake skill with ask new (using the “hello world” and “AWS Lambda” options).

Once the folder structure and files are created, I edit the .ask/ask-states.json file to reflect the information from the skill I created during the guide.

Then in the folder skill-package I remove everything except skill.json. To find what to put into that file, use the command: ask smapi get-skill-manifest -s and copy/paste that code.

Finally, I force the deploy with ask deploy --ignore-hash.

The Lambda function can now be managed locally on your computer and deployed with ASK CLI. You can go to the different skill consoles to delete the fake skill “hello world” you created.

Add a domain to a Let’s Encrypt certificate

Aymeric — Thu, 07 Jul 2022 08:03:32 +0000

For Apache, in the folder sites-available, you need to create your the-new-one.your-domain.com.conf file. Then enable the new site with a2ensite the-new-one.your-domain.com.

You can list all domains associated with a certificate:

certbot certificates

Now we add the SSL using certbot. You need to list all the existing domains and add the new one:

certbot --apache --cert-name your-domain.com -d first.your-domain.com,second.your-domain.com,third.your-domain.com,the-new-one.your-domain.com

Get email address from a Azure DevOps “by” field in Power Apps Flow

Aymeric — Fri, 13 May 2022 07:59:24 +0000

If you need to get an email from an Azure DevOps work item (e.g. from the “Changed By” field), it might be tricky in Power Apps Flow because it will return “John Doe ”.

To only extract the email from this string, you’ll have to use the below:

first(split(last(split([YOUR_FIELD],'<')),'>'))

Use CTRL and TAB to switch between two tabs in Chrome

Aymeric — Fri, 18 Mar 2022 10:38:48 +0000

It’s super handy to be able to switch between two tabs in the web browser… But it’s tricky to set it up in Chrome!

Install AutoControl: Keyboard shortcut, Mouse gesture
Install the native component as the extension asks for
Add a new action
The trigger is LEFT CTRL and TAB
The action is Switch to previous tab

It should now work to switch between tabs using LEFT CTRL + TAB.

Portable version of NodeJS on Windows

Aymeric — Sat, 01 Jan 2022 20:19:11 +0000

(This is a corrected version of this blog post)

Install Cmder in your desired location.
Download nvm-noinstall.zip from the latest release.
Extract the contents of nvm-noinstall.zip (i.e.: nvm.exe …) into the bin folder inside of the portable Cmder folder.
Navigate to the bin.
Create a new file called install_fix.cmd that contains this code.
Open a terminal in the bin folder and type .\install_fix.cmd
When asked to enter the path use the full path to your Cmder bin folder.
If it worked, a file called settings.txt has been created in the your Cmder bin folder.
Close the terminal, and reopen it in the bin folder.
Install the version of Node you want, e.g. nvm install latest.
Wait until the installation is completed, then, inside the bin folder, there should be a new folder containing the latest Node version (e.g. v17.3.0).

From there, you can use the full path to your Cmder bin + the Node version folder + node.exe to execute something with this version of Node using Powershell, CMD or other terminal.

For example:

PS D:\experiments\my-stuff\> D:\experiments\nodejs-portable\cmder_mini\bin\v17.3.0\node.exe index.js

You could use Cmder with the Node path in the Startup Settings of the app; e.g. set "PATH=D:\experiments\nodejs-portable\cmder_mini\bin\v17.3.0;%PATH%". This way, in Cmder, when typing node.exe it’s the portable version that will be used.

Redirect non-www to www on a PHPBB Forum

Aymeric — Wed, 29 Dec 2021 08:44:46 +0000

In the PHPBB directory, edit the file .htaccess and after RewriteRule ^(.*)$ app.php [QSA,L] you can enter:

RewriteCond %{HTTPS} off
# First rewrite to HTTPS:
# Don't put www. here. If it is already there it will be included, if not
# the subsequent rule will catch it.
RewriteRule .* https://%{HTTP_HOST}%{REQUEST_URI} [L,R=301]
# Now, rewrite any request to the wrong domain to use www.
RewriteCond %{HTTP_HOST} !^www\.
RewriteRule .* https://www.%{HTTP_HOST}%{REQUEST_URI} [L,R=301]

Debug a third party Android APK

Aymeric — Mon, 29 Nov 2021 15:14:42 +0000

(inspired by this blog post)

1) Install smalidea plugin

Download the smalidea plugin (see also the related Github Repository).

Open up Android Studio and you should see the welcome screen like the one on screenshot below (if not, close your current project by selecting File -> Close project), go to the Plugins section, and from the wheel icon, select Install Plugin from Disk.... Select the smalidea plugin (ZIP file) you downloaded.

2) Get the third party APK

You first need to know the type of platform where you’ll do your debug tests. To do so, make sure your device is connected to your computer (it could also be a virtual device started from the AVD Manager) with adb devices.
Then, use the command adb shell getprop ro.product.cpu.abi to find the type of processor you have. When I use my phone, I got arm64-v8a.

Go to an APK platform, like https://apkcombo.com/ and search for the Android app you want to debug. Download the APK version that fits to the type you found before:

2bis) Have a look at the APK content

You can use JADX to open the APK and have a quick look at the code.

3) Decompile APK

With APKTool, we’ll use the command: .\apktool.bat d ".\the_original_app_from_apkcombo.com.apk" -o app_to_debug.
A folder called app_to_debug is created with the decompiled version of the application.

Next, we need to copy the source files: create a folder called “src” in the new app_to_debug folder, and type cp -R smali*/* src/.

4) Import project in Android Studio

Open an existing Android Studio project and select the app_to_debug folder where you unpacked APK.

Once the project loads, you need to tell the IDE where is your source code. Make sure you’re using the “Project view” in the left side panel:

Now you can see folder structure in your left panel. Find src/ subfolder right click it and select Mark Directory as -> Sources Root.

5) Prepare App for Debugging

Open AndroidManifest.xml from the app_to_debug and find the XML element . Add the attribute android:debuggable with value “true”. Example:

6) Repack to APK

You can now repack to APK with the command .\apktool.bat b -d ".\app_to_debug\" -o app_unsigned.apk

7) Sign the APK

7a) Create a keystore

You first need a keystore using keytool and type the below command:
keytool -genkeypair -v -keystore mykey.keystore -alias mykey -keyalg RSA -keysize 2048 -validity 10000

Several questions you’ll be asked, as well as a password. Make sure to remember the password for later.

7b) Validate the APK

You then need zipalign that can be found in the Android SDK folder (e.g. C:\Users\USERNAME\AppData\Local\Android\Sdk\build-tools\31.0.0\zipalign.exe) to validate your APK:
.\Path\to\Android\Sdk\build-tools\31.0.0\zipalign.exe -f -v 4 .\app_unsigned.apk .\app_ready.apk

7c) Sign the APK

Finally you can sign the new created APK with apksigner:
.\Path\to\Android\Sdk\build-tools\31.0.0\apksigner.bat sign --ks .\mykey.keystore --ks-key-alias app_to_debug --out .\app_signed.apk .\app_ready.apk

8) Install the APK

You can install it using adb install app_signed.apk

9) Prepare the host

On your Android device, go to Settings -> Developer options and set USB debugging and Wait for debugger options on. The latter is optional but useful as it allows you wait for debugger connection and not to run app yet.

Finally, you should tap on Select debug app and choose the app you just installed. After all of these, your Developer options menu should look somewhat like this:

Now, launch the app on the Android device, and you’ll get the below message:

10) Forward debugger port

You can use the adb’s port forwarding feature and forward JDWP service where application’s debug interface is listening.

Find the JDWP port with the command adb jdwp, then use this port with the command:
adb forward tcp:5005 jdwp:JDWP_PORT

11) Connect Debugger

Go to Android Studio and from its top menu bar choose Run -> Debug…, then a small message appears with one unique option that is Edit Configurations.... There, in the window, use a plus (+) button at the opt left, and add a new configuration of type Remote. Leave the default configuration as is. Click the Debug button and your app should be running with the attached debugger which means it will stop once a breakpoint is hit and you can investigate the content of app’s variables.

Capacitor Plugin for HTTP requests with self-signed SSL certificates

Aymeric — Sat, 30 Oct 2021 13:26:01 +0000

I’m using CapacitorJS for easy development with Android. I needed a way to do an HTTPS request to a box that uses self-signed SSL certificate. To accomplish it, I created my own capacitor plugin.

See this wiki page for details: https://github.com/Aymkdn/assistant-freebox-cloud/wiki/Capacitor-Plugin-for-HTTP-requests-with-self-signed-SSL-certificates

Enable CORS with IIS

Aymeric — Wed, 13 Oct 2021 12:28:53 +0000

It’s as easy as editing the web.config file with the below: